214 109 1MB
English Pages 242 [244] Year 2018
God, Time, Infinity
Philosophische Analyse/ Philosophical Analysis
Herausgegeben von/Edited by Rafael Hüntelmann, Christian Kanzian, Uwe Meixner, Richard Schantz, Erwin Tegtmeier
Band/Volume 75
God, Time, Infinity Edited by Mirosław Szatkowski
ISBN 978-3-11-059193-4 e-ISBN (PDF) 978-3-11-059416-4 e-ISBN (EPUB) 978-3-11-059203-0 ISSN 2198-2066 Library of Congress Control Number: 2018941001. Bibliographic information published by the Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available on the Internet at http://dnb.dnb.de. © 2018 Walter de Gruyter GmbH, Berlin/Boston Printing: CPI books GmbH, Leck ♾ Printed on acid-free paper Printed in Germany www.degruyter.com
Contents Acknowledgements | VII Andrea C. Bottani and Riccardo Fedriga Ockham, Plantinga and the Row of Ants | 1 Paul Clavier The Importance of Being Timeless | 19 Christian Kanzian Temporal Relations as Epiphenomena | 39 Tomasz Kąkol In Defense of Presentism and an Extratemporal God | 53 Srećko Kovač Concepts, Space-and-Time, Metaphysics (Kant and the dialogue of John 4) | 61 Zbigniew Król Basic Intuitions Concerning the Concept of Infinity in Mathematics from a Historical and Theological Point of View | 87 Uwe Meixner No Life without Time | 105 Ludwig Neidhart God and Time. A defense of God’s timelessness | 115 Peter Øhrstrom Thoughs on Time, Truth and Transcendence | 131 Francesco Orilia The Moral Desirability of Presentism | 147 Elisa Paganini McTaggart, Lewis and the Problem of Temporary Intrinsics | 163
VI | Contents Alfredo Tomasetta Is Dualism Compatible with Classical Theism? | 171 Kazimierz Trzęsicki In What Sense is God Infinite? A Consideration from the Historical Perspective | 183 Authors of Contributed Papers | 219 Abstracts | 223 Person Index | 229 Subject Index | 233
Acknowledgements The publication of this book was preceded by the 3rd International God-Oriented Conference, with the title: God, Time, Infinity, which took place in Warsaw in September 2015. That conference was organized under the auspices of the Institute of Philosophy and Sociology of the Polish Academy of Sciences (IFiS PAN). I would like to thank the National Science Center for funding it, and also for the financial support of the publication of this book. Moreover, I am grateful to IFiS for acting as the unit implementing the project. Special thanks go to Ms. Grażyna Drążyk from IFiS PAN, who helped me in different ways throughout the preparation of the grant proposal to the National Science Center and the organization of the conference. I would like to express my sincere gratitude to all thirteen authors who accepted the invitation to contribute to the volume and showed great understanding and patience during the publication process. I would also like to thank all speakers at the conference, and Dr Bartłomiej Skowron for his faithful support of my initiatives. Last, but certainly not least, I would like to thank my brother, Sławomir Szatkowski, for his help in preparing this volume for print. I have tried to make the best possible use of all his suggestions and corrections.
Mirosław Szatkowski
https://doi.org/9783110594164-201
Andrea C. Bottani and Riccardo Fedriga
Ockham, Plantinga and the Row of Ants 1 For millennia, philosophers have discussed whether divine omniscience is compatible with human freedom – conceived of in a libertarian way – or not. If libertarianism is true, some actions are free and no action is free unless it is within the agent’s power to act otherwise. If God is omniscient, however, He completely knows how I will act in the future, which seems to entail that it is never within my power to act otherwise, provided I cannot change God’s past beliefs. Therefore, I am not free in the libertarian sense, or so it seems. In a nutshell, the problem is that every action that follows from a free choice entails a contingent view of the future, which nonetheless appears fixed to the constraint of the divine knowledge that foreknows it.¹ This problem has a long history. It was addressed, discussed and purportedly solved by a number of authors, among which Augustine, Boethius, Anselm, Aquinas, Ockham, Molina and various open theists, and has raised renewed interest after Nelson Pike’s article “Divine Omniscience and Voluntary Action” (Pike (1965)).² In his Tractatus de Praedestinatione et de Praescientia Dei respectu futurorum contingentium (see, Ockham (1969)), Ockham famously defended the compatibility of divine omniscience and human freedom by distinguishing between past propositions secundum vocem and secundum rem. In the words of Ockham himself:
1 With the term “fatalism” we refer to the idea that it is not in our power to act differently from the way we act; if we combine this idea with the notion of an intelligent, divine being which is defined by certain essential attributes (perfection, omnipotence) and, therefore, with the existence of certain propositions of a particular epistemic nature (infallibility, omniscience) that bind future states, we obtain the specific form of fatalism that goes under the label of “theological fatalism”. See, Normore (1985) and Zagzebski (2011). See also, Kane (2002); Fischer et al. (2009) and Rice (2010). 2 It is not possible to reconstruct the whole debate on compatibilism and the openness of God here: the bibliography is vast and the discussion is by no means over. For a summary of the various positions, cf., Fischer (1983); Craig (1986); Craig (1991); Zagzebski (1991); Fischer et al. (2007); Fischer (2011), pp. 165–95; Fischer et al. (2007). For a historical/theoretical reconstruction, see, Craig (1988); Marenbon (2005) and Marenbon (2013); see also, for the debate from Duns Scotus to Leibniz, Michon (2004); Mugnai (2013); Fedriga (2015). For an overview of Open Theism, see, Pinnock et al. (1994). https://doi.org/9783110594164-001
2 | Andrea C. Bottani and Riccardo Fedriga Some propositions are about the present as regards both their wording and their subject matter. Where such [propositions] are concerned, it is universally true that every true position about the present has [corresponding to it] a necessary one about the past – e.g., ‘Socrates is seated’, ‘Socrates is walking’, ‘Socrates is just’ and the like. Other propositions are about the present as regards their wording only and are equivalently about the future, since their truth depends on the truth of propositions about the future. Where such [propositions] are concerned, the rule that every true proposition about the present has [corresponding to it] a necessary one about the past is not true.³
While propositions that truly refer to the past (secundum rem, as “ Socrates was sitting” are fixed and unchangeable, others that refer to the past only verbally (secundum vocem, as “ Peter was predestined”) remain epistemically undetermined, because they are waiting to receive their proper truth-value from a state of affairs happening in the future. All foreknowledge acts fall under the latter category. Far from fixing the future, they receive their truth-value from future facts. As Nelson Pike puts it, they describe “soft” rather than “hard” facts.⁴ Soft facts about the past are not properly about the past and so they fail to be “accidentally necessary”. Since William of Sherwood (XIII century), medieval philosophers and theologians call necessitas per accidens the principle according to which, if a certain event or state of affairs is in the past, then there is nothing that one could do about it now.⁵ In common words: what’s done it is done; what has been has been. Ockham’s idea is that past divine foreknowledge, however past, is not accidentally necessary, because it is about the future. So it is within our power to act in such a way that God would not have believed what in fact he does. In Fischer’s words, Ockham’s intuition is this: Do not think of the past fact that Jones would write the paper as forcing Jones to write the paper, or constraining what Jones has it within his power to do. Rather, think of Jones’ free
3 Guillelmus de Ockham (1978a), q. I, III Propositio, ll, 208–216, p. 515 (english translation quoted in: Ockham (1969), pp. 46–47): Aliquae sunt propositiones de praesenti secundum vocem et secundum rem, et in talibus est universaliter verum quod omnis propositio de praesenti vera habet aliquam de praeterito necessariam, sicut tales: “Sortes sedet”, “Sortes ambulat”, “Sortes est iustus”, et huiusmodi. Aliquae sunt propositiones de praesenti tantum secundum vocem et sunt aequivalenter de futuro, quia earum veritas dependet ex veritate propositionum de futuro; et in talibus non est ista regula vera quod omnis propositio vera de praesenti habet aliquam de praeterito necessariam. 4 Cf., Sanders (1966); Pike (1966); Adams (1967). 5 See, Knuuttila (2013); Freddoso (1983), Maierù (1972). For the medieval notion of necessitas per accidens it occurs for the first time in a logical context in Guillelmus de Sherwood (1983), (1.7.1).
Ockham, Plantinga and the Row of Ants | 3
decision to write the paper as explaining why it was true that Jones would write the paper. In other words, says the Ockhamist, Jones’ free decision to write the paper is the explanatory ground of the fact that, at t1 , it was true that he would write the paper.⁶
Ockham’s solution faces a number of problems. Among others, both the notion of accidental necessity and the related concept of a hard fact stand in need of a more precise definition; it is far from clear what it is for an action to be within one’s power; and the purported “way out” seems to presuppose eternalism, being ultimately unavailable to those who are sympathetic to some anti-eternalist conception of time (since the “softness” of divine foreknowledge precisely consists in its dependence on the future foreknown fact, which requires the reality of the future). In his famous paper “Ockham’s Way Out” Plantinga strived to clarify accidental necessity by completely detaching this notion from the idea that the past is fixed.⁷ It is not by contrasting the closeness of the past with the openness of the future that one can get clearer about accidental necessity, because the future is just as unchangeable as the past. Consider that, in order to change the past, one should be able to do an action A at t1 such that a proposition P, referring to the past and true at t1 , becomes false at t2 , after A is done, which is impossible. But then, changing the future must be equally impossible, for nobody is able to do at time t1 an action A such that before doing A, a proposition P regarding time t2 was true and after doing A that proposition is false. Take for example the proposition “Jones will write his article tomorrow at 7 p.m.”. If the proposition is true at 3 p.m., nobody can do anything at 4 p.m. that makes the proposition false after 4 p.m. Since the future is no less unchangeable than the past, accidental necessity is not to be identified with unchangeability and has nothing to do with the asymmetry in rigidity that past and future display.⁸ Having discarded temporal asymmetry as a criterion to define accidental necessity, Plantinga manages to define the notion in terms of power of an agent. If Jones was born in 1967, there is nothing in my power that I could do today to prevent that event from happening or to modify it in any way. So, that event is neces-
6 Fischer et al. (2009), p. 256. 7 Plantinga (1986). 8 For the asymmetry of the past as the ground for transferring the strength of the necessity per accidens to the present and the future and for a solution that combines Aquinas with Ockham, Zagzebski (2011). For a historical and theoretical critique to this position, which too readily smoothes over the differences between Ockham’s Contingentist point of view and Aquinas “theological determinism”, see, Fedriga (2015), pp. 126–131.
4 | Andrea C. Bottani and Riccardo Fedriga sary per accidens.⁹ It is outside of my power, or so it seems. Here is a first tentative definition along these lines: D)
p is accidentally necessary at t if and only if p is true at t and it is not possible both that p is true at t and that there exists an agent S and an action A such that (1) S has the power at t or later to perform A, and (2) if S were to perform A at t or later, then p would have been false.
The important point is that, in the defined sense, past acts of divine foreknowledge are not accidentally necessary. Indeed, as Plantinga argues: 1) Backtracking counterfactuals of the form “If X had refrained from doing Y at t2 then God would not have believed at t1 that X would do Y at t2 ” (t1 < t2 ) are true. 2) For many actions Y and many agents X, it is within X’s power to refrain from doing Y. According to Plantinga, it is the counterfactual power of action that works as a signal of a missing accidental necessity.¹⁰ There are complications, however, and Plantinga imagines a case where a future action would be such that, were it to happen, even genuine past propositions (secundum rem, in Ockham’s words) referring to a hard fact would be false. Let us suppose that a colony of carpenter ants moved into Paul’s yard last Saturday; if the ants were to remain and Paul were to mow his lawn this afternoon, the colony would be destroyed. However, God intends that it be preserved. Paul will not mow his lawn this afternoon and God, who is omniscient, knew in advance this fact; but if he had foreknown instead that Paul would mow this afternoon, then he would have prevented the ants from moving in. So if Paul were to mow his lawn this afternoon, then the ants would not have moved in last Saturday. But it is within Paul’s power to mow this afternoon: if Paul has this power, then there is an action (mowing the lawn this afternoon) such that if he were to perform it, the proposition (E) That colony of carpenter ants moved into Paul’s yard last Saturday
would have been false. But (E) appears to describe a true hard fact, because it is strictly about the past. In the defined sense, (E) would not have necessitas per
9 Plantinga (1986), p. 247. For Plantinga’s understanding of necessity, see, Plantinga (1974). 10 Cf., Plantinga (1986), p. 253: p is accidentally necessary at t if and only if p is true at t and it is not possible both that p is true at t and that there exists an agent S and an action A such that (1) S has the power at t or later to perform A, and (2) if S were to perform A at t or later, then p would have been false.
Ockham, Plantinga and the Row of Ants | 5
accidens: Plantinga concludes that not all true propositions strictly about the past (not even hard facts) are accidentally necessary in the defined sense. Unfortunately, this kind of example can be reiterated in case of virtually all past objects and events: every individual S has the abstract power to perform an action A such that, if S were to do it, God, having foreseen it, would have refrained from creating a certain being Z, or from letting a certain event F happen. As a consequence, propositions strictly about the past (with a relation of strict implication) such as “Z existed” and “F happened” fail to be accidentally necessary and the facts that fall under accidental necessity in the defined sense become so rare that the definition seems to be hardly acceptable.¹¹ In order to solve the difficulty, Plantinga drastically restricts the actions that an agent can perform in order to prevent non-accidentally necessary past objects or events from taking place to those he calls “basic actions” – where A is a basic action just in case an agent S can perform A directly, i.e., without having to perform another action B as a means to perform A.¹² The issue is complex and controversial, so let us comment but cursorily on the difficulties faced by this proposal. A first problem is posed by the elusiveness of the notion of “basic action”. Another problem is that Plantinga is not completely clear about what he understands with our power to act so as to prevent a soft fact about the past (in particular, a divine act of foreknowledge) from taking place. As Hasker argues, to say that there are no compelling external forces preventing the agent from acting in that way is a petitio principii (it amounts to assuming that the past divine act of foreknowledge is not such a force). According to Plantinga, S has the power to act freely at t2 even if at t1 God already knows how S will act at t2 , because S has the power to act at t2 in such a way that, if S were to act in that way, God would not have believed at t1 what He does (which means that His past foreknowledge is not accidentally necessary). However, Hasker argues, if the problem is showing that divine foreknowledge does not prevent our free will (understood in the libertarian way as the capacity of acting without any constraints upon one’s will), then free will is the demonstrandum and cannot be used as a premise, which is exactly what Plantinga seems to do. Indeed, Plantinga’s argument seems to go as follows: action A of S at t2 is free because S has the power to modify God’s belief at t1 in
11 Plantinga (1986), p. 254. 12 Plantinga (1986), p. 260: An action A is a basic action for a person S if and only if there is an action A* that meets two conditions: first, S can directly perform A*, and secondly, S’s being in normal conditions and his directly performing A* is causally sufficient for his performing A.
6 | Andrea C. Bottani and Riccardo Fedriga such a way that action A is free. This would be a vicious circle, with a premise in place of the conclusion and with the explanandum used as the proof of itself.¹³ Is there any way to escape Hasker’s criticism?
2 Our aim is to defend the core of Ockham’s (and Plantinga’s) way-out by reinterpreting their compatibilism in new terms. We shall begin by arguing that God’s omniscience is incompatible with the openness of the future. Then we will show that, despite the closeness of the future, there is a fairly coherent modal sense in which it is within our power to act differently. For this modal sense to receive full articulation, a definition of hard fact is required. This is done in non-modal terms, i.e. in terms of actual grounding. So, let us begin. Suppose that: – – – –
S is a subject; A is an action; “does” is an abbreviation of “does, did or will do” (the same, mutatis mutandis, for “is true”, “knows” and the like. At t2 S does A.
Assume that: (i)
the future is open: the proposition that S does A at t can be definitely true (or false) at t2 and neither true nor false at t1 (t1 < t2 ) (ii) A proposition is a function from possible worlds and times to truth-values (and not a function from just possible worlds to truth-values). [(ii) follows from (i). If the future is open, at least some propositions must have definite truth-values at t2 that they fail to have at t1 ]. (iii) If S believes a proposition P at a time t, S’s belief is true just in case P is true at t (if I believe now that it is raining, for example, my belief is true just in case the proposition that it is raining is true now. And, if I believe now that it will rain tomorrow, my belief is true just in case the proposition that it will rain tomorrow is true now). Therefore, if at t S believes that p, then at t S knows that p just in case at t it is true that p (and, of
13 Hasker (2001), p. 103: How can Cuthbert have the power to cause Cuthbert will purchase an iguana at t3 to be false, when its truth is immutably fixed and guaranteed by the truth of God believes at t1 that Cuthbert will purchase an iguana at t3 ?
Ockham, Plantinga and the Row of Ants | 7
course S, is justified in believing that p). And, if God is omniscient, at t God knows that p just in case at t it is true that p – just in case at t p. (iv) God exists either in time or out of time (inclusive). (v) God is omniscient.
In case God exists in time, His omniscience can consist in one of two distinct abilities: (v1 ) at every time at which God exists, God knows everything that is true at that time. (v2 ) At every time at which God exists, He knows, for every time, everything that is true at that time.
Let us say that God is omniscient1 iff (v1 ) is the case, while He is omniscient2 in case (v2 ) is the case. Obviously, omniscience2 entails omniscience1 . If determinism were true, moreover, omniscience1 would entail omniscience2 , for the total set of present truths would entail the total set of future truths (as well as of past truths, if one accepts some version of a strong, bi-directional determinism). In a non-deterministic world, however, omniscience1 is not equivalent to omniscience2 but weaker than the latter. A world where the future is open is a fortiori non deterministic. Therefore, in no such world omniscience1 entails omniscience2 . On the other hand, if the omniscient God is out of time, then, for any time, He knows everything that is true at that time, but there is no time at which He knows that, so He is neither omniscient1 nor omniscient2 . Let us say that in such a case He is omniscient3 . Omniscience3 is the exact atemporal analogue of omniscience2 . With this battery of notions and assumptions in hand, we can prove what follows. 1) If the future is open, God is in time, and God is omniscient1 but not omniscient2 , then at t1 God does not know that S does A at t2 : 1) 2) 3) 4)
At t1 it is neither true nor false that at t2 S does A At t1 it is not true that at t2 S does A At t1 God does not believe that at t2 S does A At t1 God does not know that at t2 S does A
From the openness of the future From 1 From 2, modulo (iii) and (iv)1 From 3, by definition of knowledge
If at t1 God does not know that at t2 S does A, nothing in God’s knowledge prevents S from abstaining from doing A at t2 . So, if God is in time and God is omniscient1 (but not omniscient2 ), His omniscience may seem to stop threatening human freedom. But it is not clear that it does. Consider that, either at t1 it is permanently neither true nor false that at t2 S does A, or at t1 this “becomes” true
8 | Andrea C. Bottani and Riccardo Fedriga when, with the flowing of time, t2 becomes present. In the former case, nobody at t1 could make a true prediction about S’s doing A at t2 . Both the proposition that at t2 S does A and its negation are indeed permanently non-true at t1 . If X believes at t1 that S will do A at t2 , and then, when t2 becomes present, S does A at t2 , it would be incorrect to say that X’s belief was right, i.e., that X’s prediction was veridical. But this is utterly implausible. In the latter case, X’s belief is neither true nor false “before”, and “then” true at the very same time, i.e. at t1 . This kind of paradox is implied in the very idea of changing the past, and seems to entail a contradiction. The contradiction cannot be dispelled by postulating that X’s belief can be both true and non-true at t1 relative to different times (neither true nor false at t1 relative to t0 and true at t1 relative to t3 , for example), for this would close the future once again. The reason is that X’s belief is true at t1 relative to t2 just in case S does A at t2 . So, if X’s belief is true at t1 relative to t2 , S cannot refrain from doing A. 2) If the future is open, God is in time, and God is omniscient1 but not omniscient2 , then what God knows at t1 is incompatible with what He knows at t2 : 1) At t1 it is neither true nor false that at t S does A 2) At t1 God knows that it is neither true nor false that at t2 S does A 3) At t1 God knows that it is not true that at t2 S does A 4) At t2 is true that S does A 5) At t2 God knows that it is true that at t2 S does A 6) At different times God knows that it is true and it is not true that at t2 S does A
From the openness of the future From 1 and (v)1 From 2 By hypothesis From 4 and (v)1 Fom 3 and 5
3) If the future is open, God is in time, and God is omniscient2 , then God’s knowledge at a single time is incoherent. But no knowledge can be incoherent, by definition of “knowledge”: 1) 2) 3) 4)
At t2 it is true that at t2 S does A God is omniscient2 At t1 God knows that at t2 it is true that at t2 S does A At t1 God knows that it is true that at t2 it is true that at t2 S does A 5) At t1 it is neither true nor false that at t2 S does A 6) At t1 God knows that it is neither true nor false that at t2 S does A
By hypothesis By hypothesis From 1, 2 From 3 and ‘p iff it is true that p’ From the openness of the future From 5, 2
Ockham, Plantinga and the Row of Ants | 9
7) At t2 S does A iff at t2 it is true that at t2 S does A 8) At t1 God knows that it is neither true nor false that at t2 it is true that at t2 S does A 9) At t1 God knows that it is not true that at t2 it is true that at t2 S does A 10) At t1 God knows that it is true and it is not true that at t2 it is true that at t2 S does A 11) God’s knowledge at t1 is incoherent
Assumption From 6, 7 From 8 From 4, 9 From 10
4) If God exists outside of time, and God is omniscient3 , then the future cannot be open (pace Boethius). Why should one endorse 4)? The so-called Boethian solution is grounded in the idea of a tenseless God roughly as follows.¹⁴ Suppose God exists, God is outside of time and God is omniscient. Given that God is outside of time, He has no temporal states, so it is not the case that at t1 God knows that at t2 S will do some action A. Boethius argues that, if at t1 God does not know that at t2 S will do A, then at t1 S has still the power to act differently at t2 . At t1 , S can still abstain from doing A at t2 . But it is not clear that this is the case. Consider that, even if it is not the case that at t1 God knows that at t2 S will do A, certainly it is the case that at t1 it is true that God (atemporally) knows that at t2 S does A. Similarly, it is not in time that 2 + 2 = 4, for this is an atemporal truth. But since this is atemporally true, it is a fortiori true at the present time too, and at any other moment. If at t1 , and at any other moment, it is (atemporally) true that at t2 S will do A, how could S abstain from doing A in any way? If atemporally God knows that p, then atemporally it is true that p. But if atemporally it is true that p, then at any time it is equally true that p. Consider Fine’s distinction between eternal and sempiternal truths, where an eternal truth is a proposition that is true “regardless of the time” while a sempiternal truth is a proposition that is true “whatever the time”. And consider that, as Fine rightly insists, all the eternal truths (as 2 + 2 = 4) are
14 By “the so-called Boethian traditional solution” (or Boethius-Aquinas solution) we mean the thesis that all things, past, present and future, are metaphysically present to God, i.e., that God’s present is co-extensive with worldly past, present and future. In addition to this reading (see, Craig (1988) and Craig (1991)), it is worth mentioning another recent and persuasive interpretation, which we may dub “the philosophical view” (Marenbon (2013)) and is based on a close reading of the Boethian texts in their context (Cons. V, 6 ff). According to the philosophical view, the lot “all things, past, present and future are present to God”, should be understood epistemically, i.e. He knows them as if they were in His present simplicity. In the present context, we might understand this in terms of simple divine presentism in knowing things. See, Helm (2010).
10 | Andrea C. Bottani and Riccardo Fedriga a fortiori sempiternal.¹⁵ Whether the Divine knowledge that closes the future preventing S from doing otherwise is located in time or outside of it, it does not seem to make a big difference for S’s incapacity to do otherwise.
3 The moral to be drawn from 1–4 is probably that Divine Omniscience, however coinceived of, does not seem to get smoothly along with the openness of the future. So, suppose the future is closed. What would it be of S’s power to do otherwise in such a case? Is there any sense in which one might be said to be able to act otherwise in case the future is closed? Well, the future might be closed even in a nondeterministic universe. A block universe might be such that none of its temporal slice (completely) determine its subsequent slices (and each is compatible, looking in the backward direction, with more than one series of preceding slices).¹⁶ Possibility of doing otherwise might simply be logical, and physical, and chemical, and biological, and so on, compatibility of many alternative actions, with the same past. This past, however, must be purified from “soft facts”, including facts of Divine foreknowledge. This is necessary, for no two alternative actions can be both compatible with a past in which God knows that only one of them takes place. If God knows at t1 that I do A at t2 , it is true at t1 that I do A at t2 , which is (logically) incompatible with my doing B instead of A at t2 . How to give a clear-cut distinction between hard and soft facts about the past? Suppose Riccardo builds a time machine, gets on the machine, writes “500 years back” on a quadrant and press the button ‘start’. After about ten minutes the notice “here we are” appears on the quadrant just beside the phrase “500 years back”. Getting off the machine, the time traveler finds himself in Duomo Square, Milan, at noon of 1st November 1614, exactly 500 years before his departure. Call ‘A’ the traveler’s action of pressing ‘start’ and ‘F’ the machine appearance in Duomo Square at noon of 1st November 1614. If Riccardo had abstained from doing A, F would not have occurred. But F is a hard fact about the past, and it is not in Riccardo’s power to bring about that a hard fact about the past had not occurred. Therefore, it is not in Riccardo’s power to abstain from doing A.
15 Fine (2005); for the notion of Eternity, see, Helm (2010) and Kretzmann and Stump (1981). 16 By “temporal slice of the universe” we roughly mean the total state of the universe at a single time. There are of course complications with special relativity, but let us leave this aside here.
Ockham, Plantinga and the Row of Ants | 11
It is assumed that F is a hard fact about the past when Riccardo does A. In a sense, however, the arrival to Duomo Square at noon of 1st November 1614 is in Riccardo’s future when Riccardo does A (not in the subjective future of his personal time, but in the objective future of the physical time in which Riccardo is located: the physical time flowing inside the machine). It is not in question, of course, that F occurs 500 years before A: it does, but only outside the machine. Inside the machine, however, the temporal order is quite reversed, and F occurs ten minutes after A. It looks as if there were two temporal orderings. Many events, for example the Waterloo battle, have a place in the former ordering but not in the latter. Others, for example Riccardo yawning during the time journey, have a place in the latter ordering but not in the former. The only two events that have a place in both orderings are F and A, which occur in a reversed order in the two orderings. If you put the orderings together, you have a circle. No circle, of course, is an ordering, which becomes clear if one realizes that A can be seen both as a soft fact and as a hard fact about the past from any other point of the circle (the same can be said of every point of the circle: in the circle, any point is at once past and future relative to any other point). When there is anything like such a temporal and causal circle, our intuitions about which facts are “hard” and which are “soft” become hopelessly confuse. No fact seems to be “absolutely hard” or “absolutely soft”, even though they seem to be more or less hard insofar as they are more or less near in the past (the nearer in the past, the harder – just as, the less near in the past they are, the nearer in the future and so the less hard). Even in cases where such a temporal and causal circle occurs, grounding does not seem to become circular.¹⁷ The reason why F occurs, indeed, is clearly that Riccardo does A, not the reverse, even though A is subsequent to F in the “common” temporal order. So, A follows F in the common temporal order while preceding F in the order of grounding. Likewise, S’s doing an action follows in the temporal order the divine foreknowledge that S will do that action, but nevertheless it is the ground of that foreknowledge, in the plain sense in which, for any p, it is the fact that p that grounds the knowledge that p, and not the reverse (nothing can be a fact because someone knows that it is a fact; rather someone can know that it is a fact because it is a fact). So, God knows that S will do A because S will do A, and not the reverse. If this is true, one might give
17 Here and in what follows our notion of grounding is largely a non theoretical, naive one – roughly the notion of what relates, for example, the existence of a conference to the existence of its participants and their actions, or the existence of molecules to the existence of atoms, or the truth maker of a sentence to its truth. Nothing will be said here, then, on the many sophisticated questions that are nowadays widely discussed under the heading of “metaphysical grounding”.
12 | Andrea C. Bottani and Riccardo Fedriga a definition of “hard” versus “soft” in terms of grounding, regardless of temporal order, roughly along the following lines. A fact about the past is strongly hard just in case it is grounded only in the past while it is weakly hard just in case it is ultimately grounded in the past. (Accordingly, the present fact that at t3 God will know that at t1 Paul did Y is weakly hard, for it is ultimately grounded in Paul’s having done Y in the past by being directly grounded in a future divine knowledge of that action; but divine past foreknowledge is neither strongly nor weakly hard for it is grounded in future facts, so it is a soft fact). A fact about the past is accidentally necessary just in case it is strongly hard. Interpreted in modal terms, accidental necessity is a relation between worlds, while interpreted in terms of grounding it is an intra-world relation between actual facts: the latter is, at any rate, weaker than the traditional notion of accidental necessity. An action’s being within one’s power, however, remains a modal notion: Y is within X’s power at t just in case there is a world W such that X does Y in W, and the same strongly hard facts occur before t in W and in the actual world. Summing up: S is free in doing A at t iff it is in S’s power to abstain from doing A. And to abstain from doing A at t is in S’s power iff A’s abstention from doing A at t is (logically, physically, biologically, chemically and so on) compatible with t’s past, i.e., with the sum total of the world’s slices that are antecedent to t, purified from soft facts. Plantinga treats accidental necessity as a strictly modal notion: a fact about the past is accidentally necessary only in case there is no possible action that is capable of preventing it from taking place.¹⁸ On the other hand, he seems to understand the notion of a hard fact essentially in terms of strict “aboutness”: a hard fact is strictly about the past (it is, so to speak, intrinsic to the past). Moreover, he believes that some hard facts are not accidentally necessary insofar as there are some facts that are strictly about the past, yet the possibility of acting so as to prevent them from taking place obtains.¹⁹ We prefer to make of two notions one, by giving up the distinction between hard facts and accidentally necessary facts. Hard facts are nothing over and above accidentally necessary facts, and are conceived of in non-modal terms, as facts that are actually grounded in the past.
18 Plantinga (1986), p. 249. 19 Plantinga (1986), p. 254. See, the critiques of Fischer in Fischer et al. (2009), p. 261.
Ockham, Plantinga and the Row of Ants |
13
4 In order to remain true to Ockham’s approach and avoid bending his thought to fit a contemporary framework, one must safeguard three fundamental assumptions: i) ontological parsimony, i.e. the principle that one ought not to posit the actual existence of multiple and superfluous things (res)²⁰ by e.g., turning connotative terms into real beings; ii) the non-modality of temporal relations, according to which the term ‘time’ itself defines connotation²¹; and iii) the doctrine of divine simplicity, which grounds the other two. Subscribing to divine simplicity means to stick to an integrated consideration of the ontological and the theological level, just as Ockham does; but as our analysis is concerned with providing a critical reconsideration of Plantinga’s position, which separates the two levels, we opted for considering them separately too. Neo Ockhamist thinkers, like Plantinga and – from a different standpoint but in a similar methodological perspective – Hasker, seem to focus on only one of the two poles that constitute the relation between facts and acts of foreknowledge that Ockham regards as unavoidable. As a consequence of this approach, they are led to build a kind of “bridge relationships” or trans-world relationships between possible worlds, such that the ontic relatum (located in a world w1 , w2 , w3 and so on, according to necessity) can make the propositions of divine foreknowledge true. But this makes their solutions appear as dangerously ad hoc, that is, built just in order to justify the apodictic statement about human free will. In contrast with this approach, and in line with Ockham’s principles, we propose to merge the two notions into one, by treating hard facts on a par with accidentally necessary facts. Hard facts are always accidentally necessary, and are defined in terms of grounding, in a non-modal way (so that only the actual world is relevant in order to establish whether some facts are hard or not). We argue that, thus understood, Ockham’s way-out is by and large more viable than in Plantinga’s modal interpretation. Indeed, we point out that, in contrast with Plantinga, Ockham postulates a relationship of bringing about which occurs completely inside this single world, which the principle of ontological parsimony and the doctrinal grounding of faith conceive of as unique. This relationship does not occur across a number of max-
20 The principle of parsimony is strictly connected to the name of Ockham. Even if the idea can be traced to as far back as Aristotle, passing through Roger Bacon, this label perfectly fits Ockham’s thought since “his metaphysical conclusions are what we would expect from a philosopher who is assuming this principle as his guide” (Adams (1987), I, p. 157). For an externalist reading of Ockham, see also, Brower-Toland (2007). 21 Guillelmus de Ockham (1985), p. 47.
14 | Andrea C. Bottani and Riccardo Fedriga imal sums of states of affairs (i.e. possible worlds) but across different planes within the same world – which, incidentally, guarantees a more solid grounding than that granted by the unstable range of possible worlds. Thus construed, the relationship does not need a plurality of worlds; it simply moves along a single axis, where the intensity of the entailment obtaining between events is variable, depending on whether it is the ontic or the epistemic plane that is taken into account. Once the principle by which an individual S has the possibility to act without her actions being already determined is affirmed, the free act E and the free act non-E determine which relation is brought about by the implication/capacity: the one with proposition Y or the one with proposition X. In this way, the incompatibility between the freedom of human will and the necessitating foreknowledge of God is solved by moving them both into different worlds (states of affairs), in a disposition, as it were, which links the divine act of knowledge with the corresponding factual event; such a link does not constitute a simple logical possibility but rather a possible and existing state of affairs (even in the peculiar form of a possible world). This approach enabled Ockham to account for those propositions (e.g. future contingents, prophecies) that are neither true nor false as far as the reference is concerned, given that they rest on future states of affairs that are not yet closed. Such propositions, however, are determinately true according to a truth-value that rests on the premises confirmed by logica fidei (i.e. an epistemic logic of religious belief) and not on the accidental necessity of the past. In Ockham’s view, theology is the language that provides the viator (i.e the man walking this path of life) with the pragmatic rules for connecting res on the plane of ontological grounding. Those rules, nonetheless, are not to be understood as actual components of reality, insofar as they are mere connotative terms by which human language is able to understand the relationships that occur between actual things. The interest in recovering Ockham’s original solution, therefore, does not lie with the requirements of philological strictness or historical accuracy; rather, it is due to the fact that Ockham’s theory provides a clear and efficient tool for discussing the problems of theological compatibilism: namely, the notion of a unique world, the uniqueness of which guarantees the grounding’s necessity and in which events must be conceived of as poles in a relationship of variable intensity. It is in this area of Ockham’s thought that one may find an answer to the still open question concerning whether Ockham’s way-out is, in any sense, available to non-eternalists too.
Ockham, Plantinga and the Row of Ants | 15
Bibliography Ancient and medieval sources Augustine of Hippo (1887), “Sancti Aurelii Augustini opera”, in Corpus Christianorum Ecclesiasticorum Latinorum, Vindobonae: Hoelder-Pichler-Tempsky. (op. cit.: De civitate Dei libri viginti duos, PL 42, CSEL 40/1–2; De libero arbitrio libri tres, PL 32, CSEL 74; Retractationum libri duo, PL 32, CSEL 36). Anselm of Canterbury (1946–1961) Sancti Anselmi Cantuariensis Archiepiscopi “Opera omnia” ad fidem codicum recensuit F.S. Schmitt O.S.B., 6 voll, Edinburgi apud Th. Nelson et filios. Aristotle (1949), Categoriae et liber de Interpretatione, edited by L. Minio-Paluello, Oxford. See also: Aristotle’s Categories and De Interpretatione, edited by John L. Ackrill, Oxford: Clarendon Press. Aristoteles Latinus (1968), Aristoteles Latinus, IV. 1–4, edited by Lorenzo Minio-Paluello et Bernard G. Dod, Bruges-Paris. Boethius (1880), Anicius Manlius Severinus Boethius, Commentarii in librum Aristotelis Peri Hermeneias pars posterior, Hrsg. Carl Meiser, Leipzig: Teubner. Boethius (1906), Commentary on Isagoge, IIa, Hrsg. Samuel Brandt, (Corpus Scriptorum Ecclesiasticorum Latinorum 48), Vienna-Leipzig: Tempsky and Freytag. Boethius (2000), De Consolatione Philosophiae, Opuscula Philosophica, Hrsg. C. Moreschini, Munich-Leipzig: Teubner. Also: Altra ediz-consultata Anicii Manlii Severini Boethii Philosophiae Consolatio, a cura di Ludovicus Bieler, Turnhout: Brepols. Guillelmus de Ockham (1974), Guillelmi de Ockham “Summa logicae” in Opera Philosophica I, edited by Ph. Boehner, New York: St. Bonaventure University. Guillelmus de Ockham (1978a), Guillelmi de Ockham Tractatus de praedestinatione et de praescientia Dei respectu futurorum contingentium, edited by Ph. Boehner, New York: St. Bonaventure University. Guillelmus de Ockham (1978b), Guillelmi de Ockham “Expositio in librum Perihermeneias Aristotelis”, in Guillelmi de Ockham Opera Philosophica, II/2, edited by A. Gambatese and S. Brown, New York: St. Bonaventure University. Guillelmus de Ockham (1980), Guillelmi de Ockham “Quodlibeta septem”, in Opera Theologica IX, New York: St. Bonaventure University. Guillelmus de Ockham (1985), Guillelmi de Ockham “Expositio in libros physicorum”, in Opera Philosophica IV, edited by V. Richter and G. Leibold, New York: St. Bonaventure University. Guillelmus de Sherwood (1983), William of Sherwood, Introductiones in logicam, edited by Ch. H. Lohr, P. Kunze and B. Mussler, 219–99. Also in Traditio: XXXIX, 219–289. Scotus, I. D. (1963), Ioannis Duns Scoti Opera Omnia, studio et cura Commissionis Scotisticae, Vol. V, Praeside P. Carolo Balić, Civitas Vaticana. Scotus, I. D. (1966), Ioannis Duns Scoti Opera Omnia, studio et cura Commissionis Scotisticae, Vol. XVII, Praeside P. Carolo Balić, Civitas Vaticana.
16 | Andrea C. Bottani and Riccardo Fedriga Contemporary sources Adams, M. M. (1967), “Is the Existence of God a “Hard” Fact?”, in The Philosophical Review: 76, 92–503. Adams, M. M. (1987), William Ockham, Notre Dame, Indiana, (USA): University of Notre Dame Press. Alston, W. (1986), “Does God Have Beliefs?”, in Religious Studies: 22, 287–306. Brower-Toland, S. (2007), “Intuition, Externalism, and Direct Reference in Ockham”, in History of Philosophy Quarterly: 24, 317–336 Craig, W. L. (1986), “Temporal Necessity: Hard Facts/Soft Facts”, in Philosophy of Religion: 20, 65–91. Craig, W. L. (1988), The Problem of Divine Foreknowledge and Future Contingents from Aristotle to Suarez, Leiden: E. J. Brill. Craig, W. L. (1991), Divine Foreknowledge and Human Freedom. The Coherence of Theism: Omniscience, Leiden: E. J. Brill. Fine, K. (2005), “Modality and tense”, in Kit Fine, Philosophical Papers, Oxford (UK): Oxford University Press. Fedriga, R. (2015), in La sesta prosa, discussioni medievali su prescienza, libertà e contingenza, Milano: Mimesis. Fischer, J. M. (1983), “Freedom and Foreknowledge”, in The Philosophical Review: 92, 67–79. Fischer, J. M. (2011), “Foreknowledge, Freedom, and the Fixity of the Past”, in Philosophia: 39, 461–474. Fischer, J. M., Kane, R., Pereboom, D. and Vargas M. (2007), Four Views on Free Will, Oxford: Blackwell. Fischer, J. M., Todd, P. and Tognazzini, N. (2009), “Re-Reading Nelson Pike’s ’Divine Omniscience and Voluntary Action”’, in Philosophical Papers: 38(2), 247–270. Freddoso, A. J. (1983), “Accidental Necessity and Logical Determinism”, in Journal of Philosophy: 80, 257–278. Hasker, W. (2001), “The Foreknowledge Conundrum”, in International Journal for Philosophy of Religion : 50, 97–114. Helm, P. (2010), “Eternity”, in Stanford Encyclopedia of Philosophy, edited by E. N. Zalta. URL = http://plato.stanford.edu/entries/eternity/> FALL 2015 Kane, R. (ed.) (2002), The Oxford Handbook of Free Will, Oxford-New York: Oxford University Press. Knuuttila, S. (2006), “Medieval Theories of Future Contingents”, in Stanford Encyclopedia of Philosophy, edited by E. N. Zalta. URL = FALL 2015 Knuuttila, S. (2013), “Medieval Theories of Modality”, in Stanford Encyclopedia of Philosophy, edited by E. N. Zalta. URL= http://plato.stanford.edu/entries/modality-medieval/>FALL 2015 Kretzmann, N. and Stump, E. (1981), “Eternity”, in Journal of Philosophy: 79, 429–458. Maierù, A. (1972), Terminologia logica della Tarda Scolastica, Roma: Edizione dell’Ateneo. Marenbon, J. (2005), Le temps, l’eternité et la prescience de Boèce à Thomas d’Aquin, Paris: Vrin. Marenbon, J. (2013), “Divine Prescience and Contingency in Boethius’s Consolation of Philosophy”, in Rivista di storia della filosofia: 1, 9–23.
Ockham, Plantinga and the Row of Ants |
17
Michon, C. (2002), “Omniscience, liberté humaine et immutabilité divine: remarques à partir du Tractatus de Guillaume d’Ockham”, in Boulnois, Schmutz e Solère, 149–169. Michon, C. (2004), Prescience et liberté, essai de théologie philosophique sur la Providence, Paris: PUF. Mugnai, M. (2013), Possibile/Necessario, Bologna: Il Mulino. Normore, C. G. (1985), “Divine Omniscience, Omnipotence and Future Contingents: an Overview”, in Rudavsky (1985), 3–22. Ockham, (1969), William Ockham: Predestination, God’s Foreknowledge and Future Contingents, M. M. Adams and N. Kretzmann (eds.), New York: Appleton Century Crofts. Pinnock, C., Rice, R., Sanders, J., Hasker, W., Basinger, D. (1994), The Openness of God. A Biblical Challenge to the Traditional Understanding of God, Downers Grove, Illinois (USA): Inter Varsity Press, & Carslile, Cumbria (UK): The Paternoster Press. Pike, N. (1965), “Divine Omniscience and Voluntary Action”, in The Philosophical Review: LXXIV, 27–46. Pike, N. (1966), “Of God and Freedom: a Rejoinder”, in The Philosophical Review: LXXV(3), 369–379. Plantinga, A. (1974), The Nature of Necessity, Oxford: Clarendon Press. Plantinga, A. (1986), “On Ockham’s Way Out”, in Faith and Philosophy: 3(3), 235–269. Rice, H. (2010), “Fatalism”, in Stanford Encyclopedia of Philosophy, edited by E. N. Zalta. URL = http://plato.stanford.edu/entries/fatalism/FALL 2015. Rudavsky, T. (a cura di) (1985), Divine Omniscience and Omnipotence in Medieval Philosophy. Islamic, Jewish and Christian Perspectives, Dordrecht-Boston: Reidel. Sanders, J. T. (1966), “Of God and Freedom”, in Philosophical Review: 75, 219–225. Zagzebski, L. T. (1991), The Dilemma of Freedom and Foreknowledge, Oxford/New York: Oxford University Press Zagzebski, L. T. (2011), “Foreknowledge and Free Will”, in Stanford Encyclopedia of Philosophy, edited by E. N. Zalta. URL = http://plato.stanford.edu/entries/free-willforeknowledge/FALL 2015.
Paul Clavier
The Importance of Being Timeless Eternity is very long, especially toward the end.
A great amount of literature has been devoted to the relationship of God to time. Can we conceive of a timeless personal being? Or even of a timeless being at all? And, granted that the concept of a timeless being be free of contradiction, is it compatible with other attributes or attitudes traditionally ascribed to God, like omniscience, interactive dialogue with created beings, providence, foreknowledge? If, as Nelson Pike suggests, timelessness is not to be deduced from other divine attributes, on which grounds are we to discuss whether it may be ascribed to God? Another concern is the great variety of views about the nature of time, not only in philosophical debates, but even in everyday’s worldview. Because of this variety, it is not easy to get an univocal description of “timelessness”. I will assume the following: ascribing timelessness to God would not be relevant if we were adopting the thesis of “unreality of time”. If time does not exist, if there are no truly temporal relations nor properties, being timeless amounts to a tautological predicate. In order to be relevant, God’s timelessness has to be contrasted with truly temporal modes of existence. Let us grant that some beings, and in particular some agents are – at least partly – temporal beings. Let us assume that there is the A-series of time, that we experience as a passage of the time, consisting in past, present and future instants relatively to the frame of reference of a temporal observer. Let us grant that tensed propositions describe something very important about the world. I will first run through arguments pro and contra divine timelessness. I will then inquire into what a timeless God is supposed to miss. By the way, I will shortly consider some unexpected advantages of divine timelessness as regards the problem of theodicy. I will then focus on the traditional problem of compatibility of God’s omniscience and libertarian freedom, and consider to which extent God’s timelessness can solve the riddle. I will finally inquire into whether a timeless sovereign God undermines the libertarian freedom of creatures.
https://doi.org/9783110594164-002
20 | Paul Clavier
1 Eighteen short arguments for and against divine timelessness Let us first spell out some motives of denying or of ascribing timelessness to God. One reason for preferring a temporal God is the worry of keeping the highest possible amount of analogy between our current concepts and the concepts involved in various versions of theism. This can be expressed through the following argument: 1. (1) Timelessness does not make sense for us. (2) A definition of God has to make sense for us. (3) Therefore, God must not be defined as a timeless being.
Surely, the conclusion does not necessarily obtain. For it could be objecteded that God might be timeless, and nevertheless be defined in terms that make sense for us: for not every predicate convenient to X is a definition of X. Unfortunately, this reply forgets that in the case of God, there is the claim that God IS all his essential properties at once. God’s timelessness is often supposed to weaken the prior probability of theism. To be sure, framing the idea of a timeless person is a much stranger hypothesis than that of a person in time. But, following Brian Leftow¹, we could try to emphasize that, to some extent, significant parts of our moral experiences are timeless: our intuitions about the value of goodness, when we realize that such or such choice is timelessly valuable, some experiences of forgiveness and atonement (whereas some past wrongdoing is conserved but forgiven), some experience of hope (whereas we are already certain of some future happening). . . Cf., Epistle to the Hebrews 11, 2: “faith is the assurance of what is hoped for”², meaning that faith is assured that what is hoped for will become a reality, experiencing simultaneously a state of affairs as “not yet” AND the same as “already” obtained. Another kind of argument rests on the alleged likeliness of the creatures to the Creator:
1 Leftow (1991), ch. 3. 2 See, Schreiner (2015), pp. 339 ff.
The Importance of Being Timeless |
21
2. (1) The creature bears the character of the Creator (for «every agent is found to produce effects which resemble it»). (2) The creature is temporal. (3) Therefore, God is temporal.
But of course, this supposes that all the characters of the creature are to be found as such in the creator, which is obviously false, for it is clear that createdness, for instance, is not to be found at all in the creator. More formally, if x bears some likeliness to y, if x resembles y, it does not follow that they must share all their properties. Defining the similarity does not amount to: x resembles y ↔def ∀F(Fx → Fy); it is enough that under some description, x resembles y. It amounts only to: x resembles y ↔def ∃F(Fx&Fy). Anyway, on the other hand, as emphasized by Nelson Pike, “this doctrine (of timelessness) is the ultimate expression of God’s transcendence”³. Instead of undermining a philosophical construction of theism, it would be a good warrant for it. Following Aquinas via remotionis, we have to establish that “we reach whatever understanding we have of God’s attributes, by removing ‘imperfections’ that attend these qualities when possessed by finite things” ⁴. But this of course may lead to a remote God. So let us consider other arguments to God’s remote perfection: 3. (1) (2) (3) (4)
Being in time is an imperfection. Timelessness is a perfection. God is endowed with every perfection. Therefore, God is timeless.
This is not very cogent, for premiss (1) is highly questionable. Why should not “being in time” be another perfection? To this extent, a godhead endowed with every perfection would have to be both timeless and in time (enjoy temporal and atemporal dispositions). Another argument runs as follows:
3 Cf., Pike (1970), Preface p. xi. Obviously enough, there are no direct arguments to God’s timelessness from other attributes. He emphasizes that timelessness entails immutability, incorruptibility and immortality, but not conversely (idem, p. 43). Nevertheless it could be suggested that timelessness be the simplest way (or the best explanation) for being incorruptible, immortal, immutable. Timelessness makes all these attributes void of their original meaning: a timeless being cannot die, but he does not live a life which could cease anyway, a timeless being is incorruptible, for he has not even the time to be corrupted, it is immutable too, but when could he have changed anyway? He does not even keep the same as before. . . he just timelessly is. 4 Pike (1970), p. 2.
22 | Paul Clavier 4. (1) God transcends the creatures. (2) All creatures are temporal. (3) Therefore, God is not temporal.
Of course it is a weak argument, for x transcends y does not imply, but just favours the view that if y is an F, then x is not an F. We are in a symmetric case of the relationship of similarity. And once again, the relationship of transcending does not amount to: x transcends y ↔def ∀F(Fy → ¬Fx); but only to: x transcends y ↔def ∃F(Fx&¬Fy), with F designating a distinctive characteristic of x, on a metaphysically significant level, which can hardly be described better than by saying that it is the kind of characteristic exemplified by timelessness. It would be better, but more begging the question, to say: 5. (1) Existing in time, successively, is the exclusive property of created beings. (2) God is not a created being. (3) Therefore, God is timeless.
There are further arguments to timelessness, concerning the ontological status of time. 6. (1) Time does not exist apart from the world (time is a relational property or a dimension that does not exist apart from a created world). (2) God exists apart from the world. (3) God exists apart from time. (4) Therefore, God is timeless.
Or, to put it otherwise: 7. (1) Time supervenes on events (as their relational property). (2) Events occur to substances. (3) Substances depend on God the creator. (4) Time depends on God the creator. (5) Therefore, God is not submitted to time.
The Importance of Being Timeless |
23
In this view, God creates everything, including the temporal aspects of what there is.⁵ God may create at a time, (meaning that the effect of his creative operation may occur at a time) but God is not in time. Although he is operating outside time, God’s action has consequences in time. 8. (1) God possesses completely illimitable life all at once. (2) Possessing illimitable life all at once excludes temporal extension (vs. once upon a time, or sometimes, or many times or always). (3) Therefore, God is outside time.
But here the first premiss, inspired by Boethius’ definition of eternity, is begging the question, except if the clause “all at once” is revised, so that timelessness is replaced by everlastingness, infinitely extended duration. 9. (1) God is causally prior to any appearance of time. (2) Being causally prior to x entails being outside x. (3) Therefore, God is outside time.
But it might be objected that every cause is supposed to precede chronologically its effects. We need a further reason. Maybe is it provided by Augustine (Confessions XI. xiii (16)), according to whom “It is not in time that you precede times. Otherwise you would not precede all times. In the sublimity of an eternity which is always in the present, you are before all things past and transcend all things future, because they are still to come.” But is this claim an argument? Let us try to construe Augustine’s justification: 10. (1) x precedes y in time ↔def x exists at a time prior to y (x and y are part of the series of time). (2) If God preceded all times in some time, there would be a time that God does not precede. That is, a time to which he, (or at least his operating) belongs! (3) God precedes all times. (4) There is no time in which God precede all times. (5) Therefore, God timelessly precedes all times.
5 See, Plato: “Time was created along with the heaven” (Timaeus, 38b5), and Augustine: “What times existed which were not brought into being by you? Or how could they pass if they never had existence? Since, therefore, you are the cause of all times, if any time existed before you made heaven and earth, how can anyone say that you abstained from working?” (Augustine, Confessions, XI, xiii (15)).
24 | Paul Clavier But this looks more like a stipulation than like an argument. In the preceding argument, it has been alluded to God’s operation. This suggests the next argument to divine timelessness: 11. (1) God operates timelessly. (2) The operation of any being follows its nature (operari sequitur esse). (3) Therefore, God is timeless.
But of course, the two premises are not easily granted. Nevertheless I will try to advocate them in due time (if I may say in this context) hereafter. There is, too, a straightforward argument from simplicity: 12. (1) (2) (3) (5)
If S is simple, S has no temporal parts. If S has no temporal parts, S is timeless. God is simple. Therefore, God is timeless.
Some other lines of reasoning to a temporal God may sound more compelling. They emphasize the lack of relationship of a timeless God with his creatures, and suggest that a necessary condition for doing God’s job is to share the temporal dimension of the creatural world: 13. (1) (2) (3) (4) (5)
God interacts with temporal agents. Interaction with temporal agents cannot be but in time. God interacts in time. Actions conform to the nature of agents. Therefore, God is in time.
14. (1) (2) (3) (4) (5)
If God is timeless, nothing really happens to him. If nothing happens to some person, she is not a living person. God is timeless, God is not a living person. God is a living person. Therefore, God is not timeless.
15. (1) God intervenes in human history (he addresses to manhood, gives commands, answers prayers). (2) The only way to intervene in X’s history is to share X’s experience of time. (3) Therefore, God shares human experience of time.
The Importance of Being Timeless |
25
But, in order to reach the conclusion that God is in time, we need at least 2 additional premises: – –
That human experience is never timeless. And, that God cannot be timeless and share (at the same time!) the human way of experiencing time.
I will turn back later to the issue as to whether a timeless God really misses something. Some other arguments are taken from God’s alleged omniscience:⁶ 16. (1) God knows every event that is logically knowable, including contingent future actions of the free creatures. (2) All times are equally present to God’s knowledge. (3) Therefore, God is outside time (God is not subject to the A-series but has access to the B-series, through a non-temporal simultaneity).
But of course, it could be objected that the second premiss is begging the question. And alternative views of what knowing some event amounts to may give the following argument: 17. (1) (2) (3) (4) (5) (6)
God is omniscient. God knows every event. Temporal events exist in time. A true cognizer of temporal events must herself exist in time. God exists in time. Therefore,God is not timeless.
6 “All are open and laid bare to his eyes, even those things which are yet to come into existence through the free action of creatures” (Vatican Council I, Dei Filius I: DS 3003; cf., Wis 8:1; Heb 4:13). See also, Calvin, according to whom attributing foreknowledge to God means “that all things have been and perpetually remain before his eyes, so that to his knowledge nothing is future or past, but all things are present; and present in such a manner, that he does not merely conceive of them from ideas formed in his mind, as things remembered by us appear to our minds, but he holds and sees them as if actually placed before him” (Institutes of the Christian Religion, Book III, ch. 21, transl. John Allen (Philadelphia, 1813) II, 145).
26 | Paul Clavier To put it otherwise: 18. (1) (2) (3) (4)
God is timeless. A true cognizer of temporal events must herself exist in time. God does not know every event. Therefore,God is not omniscient.
Here, as quite often in metaphysics, the modus ponens of the one is the modus tollens of the other. When knowledge is supposed to involve temporal simultaneity of the cognizant and what is known, omniscience excludes timelessness. There are also considerations from scriptural suggestions or religious experiences. Not only the concept of a timeless being has to be intrinsically coherent, but it still has to fit some requirements when applied to God. The list of these requirements depends largely on the variety of denominational creeds. Think, for instance, of the afterlife which is proposed to humans by God. Is it temporal? Then it is filled with a succession of instants. It would be then relevant to claim: “Eternity is very long, especially toward the end”. But there may be alternative views of afterlife. Does it really make sense to ask: how much time did you spend in Purgatory? How long will you stay in Paradise? Or John Paul the second has just arrived in heavens, he’s not used to eternal life yet. . . Sure, it is hard to deal with timelessness. My suggestion would be, if there is an afterlife, it is likely to be a life without after. Not a second half or a bonus game. In some confessions, God’s operations seem to escape the temporal frame of reference. Let us give an example: The Blessed Virgin is believed by the Catholics to have been preserved from original sin by the anticipating grace of the Holy Spirit, in view of the merits of Christ, the Redeemer”⁷. This very strange timing is hard for a temporal being. A timeless God is more likely to manage that. There is also the issue as to whether creation is to be conceived of in terms of temporal change.
7 “We declare, pronounce, and define that the doctrine which holds that the most Blessed Virgin Mary, in the first instance of her conception, by a singular grace and privilege granted by Almighty God, in view of the merits of Jesus Christ, the Saviour of the human race, was preserved free from all stain of original sin, is a doctrine revealed by God and therefore to be believed firmly and constantly by all the faithful.” (Ineffabilis Deus, Pius IX).
The Importance of Being Timeless |
27
2 What is a timeless God really missing? Let us now turn to consider what is a timeless G od really missing. As often emphasized (Coburn, Kenny), God’s timeless existence is supposed to be inconsistent with God’s being a personal agent, genuinely interacting with human beings,which seems to be a basic requirement for an suitable concept of godhead.⁸ Let us imagine the following dialogue between a poor pious beggar and God: O Lord, What are for you Centuries? Nothing but minutes! What are for you billions? Nothing but a few cents. So please, God, give me a few cents! Okay, beggar, wait a minute!
In this dialogue, there is a misunderstanding as regards the urgency of the petitionary prayer. The petitioner needs the cash now. And God’s extratemporal answer is, supposedly, not fitting: and this unappropriateness is the nervus ridendi of the joke. Seemingly, addressing some person in a true dialogue supposes that we can share with her the meaning of temporal indexicals (“now”, “then”, “yesterday”, “from now on”, “I will be with you till the end of the world”, “Today you will be with me in my paradise”, and so on). This requires that we have a common understanding of these chronological indexicals. The main objection to God’s timelessness is then that, if God is timeless, he does not experience the coming to be nor the passing-away of things, actions and events. He can make no difference between what we call past, present, future. He has no true knowledge of temporal realities.
8 “If we are to characterize God at all, we must say that He is personal, and if personal then temporal, and if temporal then in some sense in time, not outside it” (Lucas (1989), p. 213). Cf., G. Jantzen (1983), p. ...: “A timeless and immutable God could not be personal, because he could not create or respond, perceive or act, think, remember, or do any of the other things persons do which require time. Thus, within the framework of a theology of a personal God, the doctrines of divine timelessness and immutability cannot be retained”. Well, being a personal agent does not entail necessarily being in time, although our common pattern of personal agency is associated with temporal decision. But despite the temporal feature of our moral experience, the very responsibility cannot be said to be essentially temporal. Essential to a personal agent is his causal responsibility for some deeds or operations. I suggest that the relationship between my choice or my decision and their object is not essentially temporal, even if some consequences of my decision may occur in time. First, I may discover ex post that what I have experienced is what I deeply considered as right or wrong. My moral agreement or disagreement has not forcedly to precede the experience in order to be relevant.
28 | Paul Clavier This is the bedrock of some arguments against God’s timelessness:⁹ 19. (1) (2) (3) (4)
In a genuine relationship, the related persons have to interact. Interaction requires intervals of successive instants common to both related persons. A timeless being does not exist at successive instants, it only exists. God’s timelessness precludes his partaking a succession of common instants. (even if some kind of duration can be ascribed to a timeless being, it is “an infinitely extended, pastless, futureless duration”). (5) God’s timelessness precludes genuine interaction with timely persons. (6) God genuinely interacts with timely persons. (7) Therefore, God is not timeless.
This requirement of successive instants in common looks more obvious when the interaction is conceived of in the manner of a dialogue: In a dialogue, the interlocutors have to alternate answers and queries. This alternation implies both interlocutors experiencing the same or at last overlapping intervals of time (common segments of duration). If one of the interlocutors is timeless, he does not really dialogues with the other.
As N. Wolterstorff puts it: Some of God’s actions must be understood as a response to the free actions of human beings – that what God does he sometimes does in response to what some human being does. I think this is in fact the case. And I think it follows, given that all human actions are temporal, that those actions of God which are “response” actions are temporal as well. (Wolterstorff (1975), p. 197.)
Of course both of interlocutors must share something. The questions raised must have, if not the same, at least some common meaning to the creator and the creature. But is it required that they must share the same experience of the passage of time? I do not think so. For even human interlocutors are never ensured they share the same experience of time. But, it must be recalled once more, that the timeless view of God’s knowledge is not that of a preview, that God’s knowledge of the B-series of time does not consist in a foreknowledge or in some infallible forecast. I would like to mention a famous instance of interactive dialogue with an absolutely timeless immutable God¹⁰. All along the narrative and even in the specu-
9 Cf., Stump and Kretzmann (1981), p. 45. 10 “In you it is not one thing to be and another to live: the supreme degree of being and the supreme degree of life are one and the same thing. You are being in a supreme degree and are
The Importance of Being Timeless |
29
lative sections of the Confessions, it doesn’t seem that Augustine is troubled with God’s immutability, which in no way precludes pathetic episodes of interaction, like the famous narrative of his conversion. The famous Tolle lege, a childish song encouraging Augustine to read a passage of the scriptures, and the passage of the scripture, are experienced as warnings and answers to Augustine (and Monique) distress and prayer. If God timelessly brings about these circumstances, he is not to be compared any more with the remote organizers of a treasure hunt, preparing in advance for unknown future generations riddles, trials traps and rewards. His answers and advices arrive in due time, with equal or more relevance than any attentive answer in a temporally developed dialogue. A timeless God is not forcedly an anonymous prerecorded speech-server. (Like: Please hold the line, if your query is about the meeting in Warsaw, please dial 1) Well and good, but there is still the objection as to whether God really experiences our temporal condition. God is supposedly omniscient. An omniscient being must know what it is like to experience the passage of time. Does it mean that he has to be acquainted with the passage of time? Open theists would insist: of course he has to. Eternalists would claim that it is possible to know what it is like to experience the passage of time without having to experience it. We need further considerations on the knowledge of a timeless God. There are at least three models of God’s knowledge of events: predictive, observational, practical. The predictive model is missing the point of free libertarian actions, unless we accept compatibilism, and anyway is not that of timeless God. The observational model seems to imply that God must be somehow present when a future event occurs. Surely, this epistemic access to future events is possible on Stump and Kretzmann account of ET simultaneity. But we may need an additional support for the view of an epistemic immediacy, according to which the timeless God can directly and eternally be aware of the temporal events without having to attend temporally the scene.¹¹ This support is God’s “practical knowledge”. God has a practical knowledge of every singular beings and modes, since he is the one who generates and sustains them. He knows everything that happens by bringing about the very existence and operation of every object involved in every state of affairs, including
immutable”. (Augustine Confessions, I. vi (10)). 11 This is the challenge suggested by T. D. Senor (2009), p. 85. See also, p. 74, where the challenge of a non temporal epistemic presence to temporal events is nicely defined.
30 | Paul Clavier free deeds and demands made by some creatures.¹² If we deny this view, then we get a dualistic view, according to which there is or there are objects and operations which do not depend on God. God knows what the world is like, and may even know what it is like to be a bat, a vat, a brain in vat, a heroe, a villain, an ordinary person, and so on. In order to do so, he does not need to experience perspectives on the world. In order to be truly omniscient, God must not have all the phenomenal concepts (concepts about what it is like to have such and such phenomenal experiences any finite conscious beings may have). God does not need nor want any point of view. He sees (“And God saw. . . ” that is: He sees timelessly) what he makes. Kretzmann once compared the knowledge an omniscient being has of the entire scheme of contingent events with “the knowledge you might have of a movie you had written, directed, produced, starred in and seen a thousand times. You would know its every scene in flawless detail, and you would have the length of each scene and the sequence of scenes perfectly in mind.” Nevertheless, on Kretzmann’s account, there was an objection as to whether the movie-one-man-band was really omniscient: “You know the movie immeasurably better than do the people in the theatre who are now seeing it for the first time, but they know one big thing about it you don’t know, namely, what is now going on on the screen. Thus, the similar account of omniscience regarding contingent events is drastically incomplete. An omniscient being must know not only the entire scheme of contingent events from beginning to end at once, but also at which stage of realization that scheme is now.” ¹³ But this lack of indexicality, as already emphasized, is not a lack of knowledge. The questions: where am I? and what time is it now? can be answered by a timeless being, who knows exactly the order and series of spatial and temporal locations of every being. God the generating and sustaining cause of the universe knows that because He makes that. It could be said that the timeless being, far from missing the “now”, knows all the “nows”.¹⁴
12 T. Flint (1988), p. 35, precludes the account that God knows our free actions by knowing his own intentions to cause us to act in certain ways, for such external causation is, according to the libertarian, is incompatible with freedom. 13 Kretzmann (1966), p. 414. 14 Inquiring into what kinds of things a timeless being know, wondering whether there is a timeless knowledge of what is happening now, Nelson Pike denies that statements such as “the first scene is now on the screen” and “Today is the twelfth of May” report facts that are unknown from a timeless being, for either these are not facts, or they could be referred to through statements free of temporal indexical expressions. As he puts it “all that has been established is that there are certain forms of words that a timeless individual could not use when formulating or reporting
The Importance of Being Timeless |
31
If the omniscient being is the creator, that is the generating and sustaining cause of the universe, he is actually responsible for each stage of realization. In this sense, everything is timelessly present to God, but this neither requires nor entails that God has an experience of everything happening at once. He has a nonsequential knowledge of the A-series. In this sense too, God timelessly makes everything happen in due time. As Hasker put it, “The way God knows things to be is the way things really are” does not entail “The way in which God knows things (i.e., his manner of knowing them) is the same as the way in which they exist.”¹⁵ Therefore the following view seems to me consistent: God timelessly brings about a world made of successive temporal events. God knows what he is timelessly doing. God is not committed to observe within a temporal framework what he otherwise perfectly knows by doing it. God knows perfectly what’s going on, what happens to everyone, everywhere and always.
He knows, because he does. According to Anscombe, being the first agent timelessly involved in every action, he could say “I do what happens”¹⁶. (This claim
his knowledge”. (Pike (1970), p. 95) Katherin A. Rogers complains that on Stump and Kretzmann eternalist view, or on Leftow’s “QTE (Quasi Temporal Eternity)”, “God’s supposed duration has nothing in common with the duration we know, which just is temporal extension.” (Rogers (1994), p. 14) Once again, my defence of God’s timelessness doesn’t lack the temporal extension, for God brings it about. If he does not experience it, he nevertheless makes it. No part of the A series is not due to God’s generating and sustaining power. 15 Hasker (1989), p. 166. 16 Anscombe (1957/1963), p. 53. Cf., Aquinas ST II–II (q. 33, a.1c): whereas “speculative reason only apprehends things, [. . . ] practical reason not only apprehends but also causes them”. The concept of non-observational knowledge is first compared with the knowledge one has of the position and movements of one’s limbs, that can be known “even with your eyes shut” (I, 15), and without there being any “separately describable sensations” (ibid., 13) that give rise to your knowledge. “Later on, she compares the knowledge one has of one’s actions to the knowledge of a project supervisor who directs the construction of a building from afar, without seeing or hearing any reports on its progress (ibid., 82); to one’s ability to know what one is writing even if one’s eyes are closed (ibid., 53, 82); to God’s knowledge of creation (ibid., 87); and to a list that a shopper carries with him that directs his purchases, in contrast to a list made by a detective who follows the shopper around. Anscombe (1957/1963), p. 56)´’ (Schwenkler (2015), p. 29). Of course, among the instances of practical knowledge, God’s causation of the world is the only timeless one, if we adopt the ontological relational view, rather than the transitional account of creation. So we have to justify that the former holds. In the transitional account, “once there was nothing (but God), then there was something”, creation is supposed to describe the transition from nothingness to being. Something is supposed to happen to nothing, which happens to become something. This amounts to ascribe properties to “nothing”, which is absurd.
32 | Paul Clavier sounds like a verse from the Old Testament or from the Baghavad Gita). I will further inquire into whether this practical omniscience undermines libertarian freewill.
3 The importance of being timeless in theodicy I suggest to sketch briefly one consequence of timlessness in the problem of theodicy. A temporal God is supposed to share the destiny of the world, whereas a timeless God is, supposedly, remote and indifferent, splendidly isolated. But let us notice that on both views, God is equally responsible for their permission. Nevertheless, if God is temporal, He cannot foreknowledge free libertarian contingent actions (if compatibilism is excluded). He discovers the horrendous evils that are due to free libertarian contingent actions as they go along, day after day. He has no reason to allow so much evil to occur, since He doesn’t know if the future will not be worse. Any significant evil would be enough and a good reason to stop the whole process. This view is to some extent unbearable when one turns to consider horrendous evils that God is supposed not only to allow, but actually to sustain. Given the historic accumulation of forfeits of all sorts, why, if God is observing the improvement of evil, does not He stop the process? On the timeless view, the issue is not that of the intervention, but of the timeless permission of evil. And on this point, a timeless good God is more reliable in permitting evils, than a temporal God. The former knows how the whole story turns out, then if He is a good God, he cannot have permitted something that turns bad. The latter cannot know what the free libertarian will of the creatures will produce. Evils occur in spite of him; he has no idea of what can happen due to free libertarian choices of some creatures. . . But this does not diminish his responsibility, for if God is the generating cause of the world, he remains the chief responsible of everything that occurs in the world, even of what happens through secondary causes. If, being timeless, God knows the consequences of the misuse of freewill, he may have permitted their occurrence, for the sake of greater goods like libertarian freedom, courage, moral responsibilty. But if, being temporal, God could not predict the consequences of the misuse of freewill, he ought not to have endowed his creatures with this power. As regards the problem of theodicy, the temporal view of God is not in a better position than the timeless view.
The Importance of Being Timeless |
33
4 The importance of being timeless as regard omniscience Let us now focus on the importance of timelessness as regards the problem of incompatibility between divine omniscience and libertarian freedom. Let us first recall roughly the incompatibility argument of free will and foreknowledge. I suggest a compact rephrasing of the argument: 20. (1) If at t0 prior to t, God infallibly believes that S performs a at t, it is unavoidable that S performs a at t. (2) If it is unavoidable that S performs a at t, S could not have done otherwise. (3) S freely performs a at t only if S could have done otherwise (PAP). (4) Therefore, If God infallibly believes that S performs a at t, S does not freely perform a at t.
There is a classical way out, suggested by Ockham and his interprets. The way implies that libertarian freedom consists in a counterfactual power on God’s past beliefs. Let us define this power: (CP/GPB) It was within S’s power at t to do something such that if she did it, God would have not held the belief he had at t0 prior to t
If so, the truth of the proposition describing God’s belief depends of some future event (what S will have performed at t). The proposition describing God’s belief is said to express a soft fact. But what, if God’s beliefs are timeless? Hugh Rice asserts that divine timelessness is not a good way out for escaping the incompatibility between foreknowledge and libertarian freedom. Rice considers that “the view that God is timeless would not solve the problem posed by the incompatibility argument, because there would be an equally good argument for the fixity of God’s timeless belief.” (Rice (2006), p. 134) Why? According to Rice, if God timelessly believes that some (free libertarian) agent performs some action a at t, this person must have the power so to act that God would have not timelessly believed that she would perform a at t, namely by refraining from performing a at t. Let us now define the counterfactual power of a person on God’s Timeless Beliefs: (CP/GTB) It was within S’s power at t so to act that God would have not timelessly believed that she would perform a at t, namely by refraining from performing a at t.
34 | Paul Clavier As Rice explains, we’re unable to affect the past whereas we’re able to affect the future because it is neither actual nor real. And, in his view, the power to act otherwise is linked with a power of affecting the future. But Rice seems to imply that since God’s timeless beliefs are not about the future, they are not affectable by our power to act otherwise. If God’s timeless beliefs are not about the future, they are hard facts. But it seems to me there is a flaw, for if God’s timeless beliefs are not about the future relatively to a timeless God, it does’nt follow that they are past or fix. They are just timeless. God’s timeless belief that John performs freely a at t does not entail that it is fixed at t0 prior to t that God believes that John will perform freely a at t. There is no need to make the truth of God’s belief depend on a future event. For the event is future relatively to the temporal framework of S, but not relatively to God. To put it briefly, there is no future for timeless beliefs. As often emphasized by Eleonore Stump, God’s infallible knowledge is not a foreknowledge which undermines the exercise of incompatibilist freewill (I mean freewill with the power to act otherwise). To this extent, as Kretzmann and Stump put it, “the short answer to the question whether God can foreknow contingent events is no”. God’s atemporal omniscience is in no way a foreknowledge, for his beliefs are not temporal. Let us elaborate more carefully the point: God’s belief at t0 prior to t that S will freely perform a at t is no doubt a belief about the future. If S, as a non-compatibilist free libertarian agent, is able to affect the future, God’s belief about what S performs in the future depends on S. But God’s timeless belief that S will freely perform a at t is not about the future relatively to God. Then the conditional “if S had done otherwise, God would have not timelessly believed that she would perform a at t” does not obtain.¹⁷
17 To this extent, it could be said that literally, God never believes p, meaning that there is no t at which God’s belief takes place. Of course, it is true at every t that God timelessly believes that p. . . So to say, it is always true that God believes p timelessly, and not true that God always believes p. Following this line of reasoning, we are not committed to deny the principle of the necessity of the past. The beliefs of a timeless God are not bound by this principle. Not that God can bring about any change in his past beliefs. But precisely, a timeless God has no past beliefs. He timelessly has beliefs about events, some of them are past from our temporal frame of reference. And it is pointless to talk any person having the power so to act that God would have not believed that they act this way. For a true timeless belief cannot be revised.
The Importance of Being Timeless |
35
Identifying libertarian freedom with a power to affect the future, and then to make something such that if one did it, God would have not held the belief he had, this may be legitimate, as long as God’s belief are temporal. But it is precisely what is at stake. If God’s belief are timeless, the fact that they are not affectable by S does not make those beliefs fix. Of course it would be problematic that God’s infallible (timeless) knowledge of free actions be revealed before they occur. Suppose God (timelessly) reveals to some prophet at t0 that John performs a at t. (This is known as the problem of the possible prophet). In this the case, John’s performance at t is a either a determinist action, or a case of compatibilism. But, as a matter of fact, does it occur? Do we need at all cost to warn off the spectre of divine timeless beliefs rendered fix by their revelation in some time? Let us consider briefly some examples. Neither Judas betrayal, nor Peter’s denial have to be considered like free libertarian incompatibilist acts. Peter is precisely predicted that he will deny three times, but that when he has recovered his mind, he will be able to strengthen his brothers (Luke, 22: 31–34, and cf., 22: 22: “And truly the Son of man goeth, as it was determined”). And the acceptance of the blessed Virgin could be a case of compatibilist answer (is not she filled with divine grace). The possible prophet is a conceptual problem, but there may be no occurrence, from the part of God, of one’s free action been revealed to some person so that she couldn’t act otherwise and nevertheless would be supposed to act freely. So, contrarily to what Rice suggests, God’s timelessness does not have the consequences he seems to imply in terms of God’s beliefs being fix¹⁸. Are we then committed to conceive of divine timeless beliefs in terms of soft facts? Are we committed to soft eternity? But how could a timeless belief depend on a temporal fact, like John’s performance of a at t? I will suggest that John’s free libertarian performance at t depends on God’s timeless belief, and not conversely. God’s practical knowledge may account for his timeless omniscience. But of course, as we have seen in considering God’s practical knowledge, there is still the issue as to whether it is consistent with libertarian freedom.
18 As Rice (2006), p. 139, puts it: “the incompatibility argument would at best provide additional support for one’s belief that God is timeless. It would not provide an independent reason for it.” I will try to search independent reasons for God’s timelessness.
36 | Paul Clavier
5 Final remark I have suggested that the radical ontological dependence of the world on God’s creating operation accounts for his perfect knowledge of what happens: he knows what he makes. It can be objected that the cost of this view is the lack of autonomy of the creature. My last point will be to argue against this objection. What does the statement “God brings about that X plays the piano” amount to? Well, we should explain: God says “Let X play the piano” and X played the piano. And it was good (suppose X = Wladyslsaw Szpilman). Now if you ask whether Szpilman plays the piano himself, you will be answered: yes he does. If you insist, by asking whether he plays alone, the answer is still: yes. None else, at the moment, plays the piano but Szpilman. And nevertheless, ultimately, God makes Szpilman doing it alone. God does not play the piano instead of Szpilman. But God remains the Necessary and Sufficient Condition of Szpilman’s playing the piano. God is the primary cause of X’s playing the piano¹⁹. No part of Szpilman’s playing the piano is not due to God; but God’s making Szpilman play the piano implies that Szpilman, not God, plays the piano. Szpilman may be freely playing the piano, or contrained by a nazi officer, this does not matter. The kind of operation we ascribe to God when we say he makes Szpilman play the piano is specifically different from the kind of actions or reactions we ascribe to the piano-player. It is not the same issue, and the term “making”, “maker” have not the same meaning when applied to a timeless creative and sustaining principle, as when applied to a temporal created and sustained thing. Furthermore, this operation is still utterly different from the action we normally describe with factive verbs. The meaning of the factive verb in the sentence “God makes Szpilman play the piano” is not of the same kind than, say, in the sentence: “Szpilman makes you weep”. The difference is not merely the difference between accidentally and essentially ordered causes. Szpilman could be an essentially ordered cause of your
19 Cf.,“For it is God who works (who is at work) in you to will and to act in order to fulfill his good purpose.” (Epistle to the Philippians 2: 13) We advocate here a classical theistic view, following Aquinas: “And thus God may be said to be the cause of an action by both causing and upholding the natural power of the being. He gives everything the power to act, and preserves it in being and applies it to action, and inasmuch as by his power every other power acts. And if we add to this that God is his own power, and that he is in all things not as part of their essence but as upholding them in their being, we shall conclude that he acts in every agent immediately, without prejudice to the action of the will and of nature.” (De Potentia, 3, 7, corpus). But we intend to emphasize that this view makes the paranoid fear of strong concurrentism needless.
The Importance of Being Timeless |
37
laughing, being continually the one who makes you weep, without being the ultimate agent of your being weeping. The making of Szpilman playing the piano does not occur in this world, contrarily to his (eventually) making you weep (and also to the playing of the piano). The causal dependence upon God does not interfere with the causal contribution of secondary causes. They are not causes of the same kind: the first operates timelessly, the second in time. Talk of “primary” and “secondary” causes is misleading, for it seems to imply that the former differ from the latter only in size and time. But a timeless God is not chronologically prior, or posterior to what happens; he has not to be temporally simultaneous with any state of affairs he would then had to observe. He is the timeless Maker of a temporal world, knowing exactly what he makes. He does not overdetermine nor destroy freedom. To conclude: it is not easy to assess the balance of the arguments pro and contra God’s timelessness. The temporalist view is supposed to provide us with an analogical concept of divine person, whereas divine timelessness makes God lacking an essential feature of godhead such as being a true living person interacting with his creatures in a genuine dialogue. I have suggested to emphasize the threefold importance of being timeless: 1°. It solves the problem of incompatibility between divine foreknowledge and libertarian freedom better than the temporalist view, by cancelling the concept of foreknowledge; 2°. God’s timeless omniscience has not the scandalous consequences in theodicy it is often associated with; and 3°. God’s timeless omniscience does not undermine, but underpins the autonomy of creatures.
Bibliography Anscombe, G. E. M. (1957/1963), Intention, Oxford: Basil Blackwell. Anselm (1998), “The Monologion”, translated by S. Harrison, in Anselm of Canterbury, The Major Works, edited with an Introduction by B. Davies and G. R. Evans, Oxford: Oxford University Press. Augustine (1991), The Confessions, translated by Henry Chadwick, Oxford: Oxford University Press. Boethius (1969), The Consolation of Philosophy, translated by V. E. Watts, London: Penguin Books. Burrell, D. (1986), Knowing the Unknowable God, Notre Dame: University of Notre Dame Press. Calvin, J. (1960), Institutes of the Christian Religion, kedited by John T. McNeill, translated by F. L. Battles, London: S.C.M.. Craig, W. (1998), “The Tensed vs. Tenseless Theory of Time: A Watershed for the Conception of Divine Eternity”, in Questions of Time and Tense, edited by R. Le Poidevin, Oxford: Clarendon Press, 221–250.
38 | Paul Clavier Craig, W. (2001), Time and Eternity: Exploring God’s Relationship to Time, Wheaton: Ill. Crossway Books. Cross, R. (2005), Duns Scotus on God, Aldershot: Hants., Ashgate. DeWeese, G. J. (2004), God and the Nature of Time, Aldershot: Hants., Ashgate. Flint, T. (1988), Divine Providence: The Molinist Account, Cornell University Press. Ganssle, G. E. and Woodruff, D. M., eds., (2002), God and Time, Oxford: Oxford University Press. Halbertal, M. and Margalit, A. (1992), Idolatry, Cambridge: Mass. Harvard University Press. Hasker, W. (1989), God, Time, and Knowledge, Ithaca, N.Y.: Cornell University Press. Helm, P. (1988), Eternal God, Oxford: Clarendon Press. Jantzen, G. (1983) “Time and Timelessness”, in A New Dictionary of Christianity, edited by A. Richardson and J. Bowden, London: SCM. Kenny, A. (1979), The God of the Philosophers, Oxford: Clarendon Press. Kretzmann, N. (1966), “Omniscience and Immutability”, in Journal of Philosophy: 63(14), 409–421. Leftow, B. (1991), Time and Eternity, Ithaca: Cornell University Press. Lucas, J. R. (1989), The Future, Oxford: Blackwell. McTaggart, J. M. E. (1908), “The Unreality of Time”, in Mind: 17, 456–473. Maimonides, M. (1963), The Guide of the Perplexed, translated by S. Pines, Chicago: The University of Chicago Press. Pike, N. (1970), God and Timelessness, London: Routledge and Kegan Paul. Rogers, K. A. (1994), “Eternity Has No Duration”, in Religious Studies: 30, 1–16. Rice, H. (2006), “Divine Omniscience, Timelessness and the Power to do Otherwise”, in Religious Studies: 42, 123–139. Schreiner, T. R. (2015) Commentary on Hebrews, Nashville: B&H Publishing Group. Schwenkler, J. (2015), “Understanding ‘practical knowledge”’, in Philosopher’s Imprint: 15, 1–32. Senor, T. D. (2009), “The real Presence of an eternal God”, in Metaphysics and God, Essays in Honor of Eleonor Stump, edited by K. Timpe, New York and London: Taylor and Routledge. Stump, E. and Kretzmann, N. (1981), “Eternity”, in Journal of Philosophy: 78(8), 429–458. Swinburne, R. (1977), The Coherence of Theism, Oxford: Clarendon Press. Swinburne, R. (1994), The Christian God, Oxford: Clarendon Press. Wolterstorff, N. (1975), “God Everlasting”, in God and the Good: Essays in Honor of Henry Stob, edited by C. Orlebeke and L. Smedes, Grand Rapids, Mich. Eerdmans, 181–203.
Christian Kanzian
Temporal Relations as Epiphenomena The title of our volume is “God – Time – Infinity”. My article is mainly dedicated to the second item, namely: time. My aim here is to sketch a theory of time – or more specifically of temporal relations, since I am primarily interested in McTaggert’s B-series. My approach is ontological, in that my theory about temporal relations is embedded in the context of a specific framework of ontological categories. To make a long story short, I contend that temporal relations are epiphenomena, constituted by events. In order to make this thesis and its implications plausible, it is necessary to present a framework of categories in which events have their proper place (section 1). I next attempt to clarify how events could be the ontological basis of temporal relations (section 2). My next step is to address what it means to say that temporal relations are constituted by events. A key idea is that constitution is a formal relation which implies the epiphenomenal status of what is constituted (section 3). Finally, I aim to apply my account to God’s relation to time and to the relation between infinity and time (section 4).
1 The ontological framework My presentation of the ontological framework in which I am working necessarily lacks specificity. It is a sketch, but it nevertheless includes all of the categories that I regard as the basic elements of reality. Thus, I assume that there are substances or, as I call them, things; I also assume modes, and so-called occurrences, which include states and events. Moreover, all of these elements of being are particulars: concrete (i.e., non-abstract) individuals (i.e., non-universals). The central category of my ontology is that of things, the most important ontological feature of which is three-dimensionality. Things have three spatial dimensions but no temporal one. They are not extended in time but only in space: they have no temporal parts, but only spatial ones. This implies that they are wholly present at each moment of their existence. They are in a strict sense identical with themselves not only synchronically, at each moment, but also throughout time (or diachronically), and across possible worlds (or modally). They are endurers, in
https://doi.org/9783110594164-003
40 | Christian Kanzian David Lewis’s sense.¹ Things exist independently in a specific and unique sense. This is why they may be called the primary entities. However, they will not be the focus – at least not explicitly – of this article. As endurers, things are the bearers of properties. Things are not bundles of properties, for their nature does not consist of properties. They are characterized by properties. Moreover, being so characterized belongs to the nature of things. The properties that characterize things are, ontologically speaking, modes. Modes, which are my second category, are individual qualitative “determinates”, such as this brown of this table. They are strictly or existentially dependent on the things which they characterize. This brown cannot exist without this table.² Modes will play only a minor part in my theory, but they must be mentioned to convey a sense of the ontological composition of the third category: occurences, which comprise states and events. Modes, in virtue of characterizing things, are parts of the composition of states. This brown, of course together with the table, composes the state of the table’s being brown. If along with static modes like colour- or mass-determinates we also admit dynamic modes, which can be analysed as alterations of non-dynamic ones, we can assume that such dynamic modes – together with their bearers – compose entities that are ontologically very close to states: namely, events. I assume that states and events are ontologically so similar that we may subsume them under the category of occurrences. All occurrences have ontological features that clearly distinguish them from things, most notably that they are spread out in time as well as space. Occurrences must also be distinguished from modes, which I will not discuss here. As I said above, occurrences, particularly events, will play the largest role in the present theoretical enterprise, especially insofar as they serve as the ontological basis of temporal relations.
2 Events as the ontological basis of temporal relations The key to understanding events as the basis of temporal relations lies in the specific four-dimensionality of events. Both states and events are four-dimensional entities, but we can safely ignore states in this discussion. So the remaining question is how the four-dimensionality of events is best understood.
1 Cf. Lewis (1986), p. 202. 2 I use “mode” in the sense of Lowe (2006) or Heil (2003).
Temporal Relations as Epiphenomena |
41
We can start by examining a seemingly uncontroversial statement by Jonathan Lowe, who writes that “time necessarily involves change – by which I mean that time necessarily involves happenings or events.”³ According to Lowe, events are necessary for time and temporal relations. Without events, there is no time – or, in broader terms, in a static cosmos there are no temporal relations. If we accept this fundamental thesis, we can plausibly assume that the direction of time will also have to do with events and their internal structure, where the latter can be understood as the succession of different parts or phases. earlier b.
p.2
later p.3
p.4
p.5
p.6
p.7
e.
temporal relations ↑ event-phases
Events have a beginning (b.), or a first phase, which is succeeded by some other phases (p. 2–7), until the events come to an end in the last phase (e.). Temporal relations, of which the relation earlier/later is paradigmatic, seem to be immediately derived from this succession of event-parts or -phases. The beginning is earlier than, and the ending later than the middle-parts of an event. It is outside the scope of this article to apply the earlier/later relation to a reconstruction of other timeindicating concepts like “temporal overlapping” or “simultaneity”. The present aims are limited to discussing, with reference to this scheme, the way in which the temporal dimension of four-dimensional events manifests itself: events have an inner structure from which temporal relations are immediately derived. Our speaking about the fourth dimension of events refers to no more than this.⁴ In the next section we will take a more granular look at this immediate derivation of temporal relations from the succession of event-phases. Constitution as a formal relation or tie will, as mentioned above, play the decisive role.
3 Constitution: the formal tie between events and temporal relations This section will show temporal relations to be constituted by events. In order to fulfill this ontological function, constitution must not be understood by itself as a
3 Lowe (1998), p. 121. 4 I must neglect the three spatial dimensions of events, to come to my interpretation of the forth, the temporal one.
42 | Christian Kanzian dyadic entity. For if it were, it would be something genuinely ontological linking events and temporal relations, which would contradict the postulated immediate connection between events and time. Moreover, constitution must be assumed to be a relational tie, which generates a specific dependence of temporal relations on events. We must accept this if Lowe’s implication is to have an ontological basis. Last but not least, the specific constitution-dependence must be shown to be consistent with the epiphenomenal status of temporal relations. In order to capture these functions of constitution, we can introduce it as a particular kind of formal relation or tie. I do not intend to present a full ontology of formal relations;⁵ I will instead restrict myself to the minimum that is necessary to discuss constitution as a formal tie that links that which is constituted with that which does the constituting, by way of an epiphenomena-specific ontological dependence relation.
3.1 Formal relations We take our first important hint about formal relations from Kevin Mulligan’s conception of internal or, as he calls them, thin relations. According to Mulligan: “. . . a relation is internal with respect to objects a, b, c etc., just if, given a, b, c etc., the relation must hold between and of these objects”.⁶ This means that the existence of some objects is sufficient and necessary for the relations in question. Jonathan Lowe, who calls Mulligan’s internal relation “grounded relations”, regards them as “entirely determined by their relata”. They offer “no additions to reality”, on Lowe’s account.⁷ Lowe’s “no addition to reality” statement means that relations which are a) necessarily and sufficiently given with the existence of some objects, and b) entirely determined by these objects, are themselves no entities or elements of beings. There is a wide range of relations that fall under this initial definition of internal relations; one example is the bigger/smaller relation. Another is the relation of characterization, which occurs between a mode, a particular property, and a thing which the mode characterizes. For bigger/smaller this is clear: given an object x with size F, and an object y (not numerically identical with x) with size G (not qualitatively identical with F), the bigger/smaller dyad is completely deter-
5 For more details concerning formal ties see my article “Existential Dependence and other Formal Relations” in: Szatkowski (2015). 6 Mulligan (1998), p. 344. 7 Lowe (2006), p. 46.
Temporal Relations as Epiphenomena |
43
mined and therefore grounded in x being F and y being G. Applying Mulligan’s and Lowe’s rule, bigger/smaller makes no contribution to reality as an entity in itself. An analogous point holds for characterization. The choice of my examples (bigger/smaller, and characterization) should indicate both the scope and the diversity of the relations at issue. This raises the question: Shouldn’t we assume different groups among our non-entity-relations? Lowe presents a criterion for distinguishing different types of grounded relations; his criterion is given by the observation that those relations whose occurrence is due to the nature or the ontological form of their founding instances may be distinguished from those for which this is not the case.⁸ Take characterization for instance. If a mode F characterizes an object x, it combines with x because of what it is (namely, a mode), and it does so due to its nature or form, which consists in being a way in which a thing is. I suggest calling such internal relations formal. Take as a contrary example bigger/smaller. That an object x is bigger than an object y has nothing to do with its nature or its form, but is rather accidental for both x and y. Let us call them thin relations in Mulligan’s sense. Lowe has another, more metaphorical, explanation of this distinction between formal and thin internal relations: for the duration of the formal relation, the relata are “made for each other”.⁹ I would like to add a non-metaphorical distinguishing mark: thin relations seem to be in a proper sense derived from accidental aspects of their founding instances. X’s being bigger than y is derived from x’s (accidentally) being F and y’s (accidentally) being G. For formal internal relations, by contrast, we cannot allow this accidental one-way-derivation. Since formal relations concern the nature or the form of their relata, they are by definition ontologically indispensable to those relata. This is self-evidently the case with characterization. The characterization-function of a mode F is not derived from an “accidental” aspect of F, but rather is fundamental to it. Formal relations, to draw once again on Jonathan Lowe, are not derivative, but “too fundamental ... to be something in the world – an element of being – because it is that without [which] there could be no beings and so no world.”¹⁰ Let me add another important point that holds for formal, but not for thin, relations: Formal relations have founded some sort of ontological dependence between their relata. Mulligan and Lowe consider this aspect too. Mulligan speaks of an “involvement” which concerns all formal relations,¹¹ and Lowe describes
8 Lowe (2006), p. 48. 9 Ibid., p. 47. 10 Ibid., p. 49. 11 Cf. Mulligan (1998), p. 345.
44 | Christian Kanzian the co-occurrence in question as something to be apprehended from the other side, that is, from dependence: “... all dependence relations are, in a certain sense, founded upon ... formal relations – relations which are, for this reason, ontologically more basic than the dependence relations themselves.”¹² Space does not permit taking sides between these two approaches. Rather, we are highlighting the constitution-relation, and want to ask how defining constitution as such an internal, formal, dependence-founding tie may lend support to our claim that temporal relations are epiphenomena.
3.2 Constitution as formal relation Before giving an account of constitution as a formal relation in the sense introduced above, let us avoid misunderstanding by a brief sojourn along the via negativa, saying what constitution is not. To put it roughly: we are not using “constitution” in the meaning that, for instance, is given by Lynne Rudder-Baker. According to her: “[t]he fundamental idea of constitution is this: when a thing of one primary kind is in certain circumstances, a thing of another primary kind – a new thing, with new causal powers, comes to exist.”¹³ The basic idea of constitution, according to Rudder-Baker, is that constitution is a relation that occurs between things which emanate on different levels of reality, where “thing” is understood as a technical ontological term comparable to the sense introduced above.¹⁴ According to this sense of constitution, a thing of a primary kind F becomes a thing of a primary kind G (which is different of F). The constituted thing (of kind G), y, cannot be the same as the constituting one (of kind F), x. Emanation brings about something new. Rudder-Baker uses the example of the relation between a statue and its stuff or its material. According to Rudder-Baker, a lump of bronze constitutes, under some suitable circumstances, the statue – that is, it lets the statue emanate from the bronze. But the constitution-relation does not occur only between lumps of matter and macro-things; it also occurs at other levels of reality, from the very “top” level to the very “bottom” levels of micro-physics, such as parcels of mass, atoms, quarks, etc. The view developed in this paper is constructive, not polemical, and for this reason we will not object extensively with Rudder-Baker’s concept of constitution.
12 Lowe (2006), p. 34. 13 Rudder-Baker (2007), p. 32. 14 Cf. Rudder-Baker (2007), pp. 33ff.
Temporal Relations as Epiphenomena |
45
I just want to mention in passing that the multi-layer-picture of the world implied by Rudder-Baker’s constitution-theory lays itself open to a number of objections.¹⁵ One is that, according to my ontological scheme, lumps of bronze or other materials cannot be things in the sense introduced above, but rather quasi-individuals with indeterminate identity.¹⁶ It is even less plausible to regard micro-phenomena like quarks and atoms as things – but that is another story. To return to our theme of the theory of constitution: how can we understand it in a non-Rudder-Bakerian way with the tools I have already introduced? Three points can be made, the first two of which are implicitly non-Rudder-Bakerian. The third will mark the difference between my approach and Rudder-Baker’s explicitly. First, constitution is an internal relation and not an entity. The standard argument here is as follows: if constitution were an entity, what would relate it with the constituting and the constituted? Another, second-level relation? Should we regard these second-level relations as further entities? It is obvious that no matter how far back we go, the regress will extend. On the other hand, should we assume the second-level relations to be non-entities? This would stop the regress, but how could we argue against the entity-status of the second-level relation, if we accepted it for the first-level relation? Constitution is internal, but, second, it is not a thin relation, but rather a formal one. Whatever constitution may be, it pertains to the nature of the constituted to be constituted. It is not accidental to it. The same holds for the constituting. Nothing can constitute something due to an accidental aspect. Our third point is that the logical-formal character of constitution can at best be understood with reference to the dependence that is typically founded upon formal relations. This constitution-specific dependence relation has three characteristics. (i) It is irreflexive: nothing can constitute itself and thus create an autodependent founded entity. With regard to the problems with Rudder-Baker’s concept of constitution, we can take this irreflexivity not only as an individual but also as a generic affair: no x of a kind or category F can constitute another y of the same F. (ii) It is asymmetric: if y is constituted by x, x cannot be constituted by y. No two entities can mutually stand in constitution-founded dependence. Here, too, this condition has a generic dimension: if an x of kind F constitutes a y of kind G, it is impossible that another G constitute x or another instance of F. Finally (iii), it is non-transitive. This distinguishes our concept of constitution definitively from
15 I think John Heil’s arguments against a “multi-layer-picture” particularly convincing; see, Heil (2003), pp. 28f. 16 In the sense of what Lowe calls “quasi-objects”, see e.g. Lowe (1998), p. 58.
46 | Christian Kanzian Rudder-Baker’s. If an x (of kind F) constitutes y (of kind G, not identical with F), neither y (nor another G) can constitute some z, which would then also be constituted by x. The dependence of the constituted y on the constituting x deprives y of being the constituent, and thus of being the dependence-basis, of some other z. Incidentally, this blocks any possibility of “multi-layer-models” of reality. For this reason, all that is constituted must be understood as something ontologically secondary, as an epiphenomenon. Any obscurity in this conclusion will be made explicit in the next section.
3.3 Events and epiphenomenal temporal relations We now have the theoretical elements in hand to explore the way in which time and temporal relations are constituted by events. Let me start by characterizing the constituting tie between events and temporal relations as internal. It follows that the constitution of temporal relations by events (where events are understood as unified sequences of continuously occurring parts or phases), is not an entity unto itself. There is no dyadic entity that exists between a sequence of event-phases and the temporal relations they constitute. It is the occurrence of sequences (or phases) that constitutes e.g. being earlier or being later. This is how the assumed immediacy between events and time, according to which events are paradigmatic temporal entities, are to be explicated. This explication precludes understanding the constitution of time by events in terms of adding something to reality. The constitution of temporal relations by event-phase-sequences is formal internal, as opposed to thin internal, in the sense introduced above. It is due to the nature or the form of events, which are phase-sequences, that they constitute temporal relations. And it is in the “nature” of temporal relations to be constituted by events. If the constitution of temporal relations by events is formal internal, then it is also – in accordance with the foregoing – a dependence-founding relational tie. That temporal relations are constituted by phase-sequences yields the result that temporal relations are ontologically dependent on phase-sequences. With our formal tools we can describe this kind of dependence so as to legitimate an antirealistic or epiphenomenal account of the constituted. First, we can state that the dependence between temporal relations and eventphases is irreflexive, since reflexivity can, for every constitutional tie, be excluded at both the individual and the generic levels. An event can be tied neither to itself nor to another event in virtue of the way which it is tied to any temporal relation. And nor can a temporal relation, in virtue of how it is connected with an event, be
Temporal Relations as Epiphenomena |
47
tied to itself or to other temporal relations. These observations are intuitively plausible. Moreover, their ontological implications are relatively unsurprising. Just to mention one such implication: if the suggested sort of irreflexivity applies, then no time can be an event, and no event a time. A realist about time must establish time as a category of being unto itself. But the strategy of construing time as a sub-kind of event-like entities, as we have seen, is blocked by our construal of constitution – which, although it does not preclude a realistic account of time, makes it less plausible. The second formal aspect of constitution is individual and generic asymmetry. When events constitute temporal relations, the resulting dependence of temporal relations on events precludes this dependence from occurring between events and temporal relations. Moreover, the asymmetry of dependence brings us closer to an anti-realistic account of temporal relations, and consequently to an epiphenomenal interpretation of time. If events do not depend on time in an ontologically significant way, then what else does? If nothing depends ontologically on time, then what could render the very idea of being an entity, or the existence of temporal relations, intelligible? For it is downright unintelligible to deny that entities ground ontological dependence in at least some way. This consequence is also supported by the third formal characteristic of constitution. It is non-transitive. If temporal relations are constituted by events, then such relations cannot constitute anything else that would have been regarded as the product of the constitution of events. Being constituted by events renders it impossible for temporal relations to serve as the constitutional basis of some other relata. The conclusion of the foregoing is that, since constitution is an internal relation, there is no entity “between” events and time. Constitution is a formal internal. The reason is that events consist, essentially or “in their nature”, of sequences of phases, which constitute time. This formal tie between events and temporal relations brings the latter into a kind of dependence on the former; this makes it impossible for temporal relations, themselves, to be the basis of constitution. It is especially important to note that, since time cannot serve to ground any other form of ontological dependence, the alternative arises of considering time to be a sort of epiphenomenon. Standardly, because epiphenomena are phenomenally and, to some extent, theoretically indispensable, they are considered ineliminable even though they play no role in the causal nexus of the world. Some authors regard mental phenomena like qualia as epiphenomena. I do not intend to contribute to this debate. I shall merely suggest – with reference to time – that we regard time as indispensable to any theory of the world, but that we shift our focus from the causal to the constitutional inefficacy of epiphenomena. Because
48 | Christian Kanzian time and temporal relations are epiphenomena, they are indispensable products of constitution, but they cannot be the basis of some further constitution, nor can they ground any other kind of ontological dependence. If epiphenomena are inefficacious, then, in spite of their (phenomenal) indispensability, they cannot be considered to belong to the basic structure of reality. Epiphenomena are not entities. This result fits well with the standard analytic understanding of epiphenomena. In our case, this would imply a kind of moderate anti-realism about time and temporal relations. Constitution as a formal relation can be re-interpreted from this standpoint as well: Constitution is that formal internal tie which grounds a strong enough ontological dependence so that the dependent thing loses the status of entity and must thus be regarded as epiphenomena. In short: constitution ties entities together with epiphenomena. Events are entities which – being unified sequences of different phases – constitute epiphenomenal temporal relations. It would be interesting to consider additional arguments for this claim, especially from the perspective of a thing- or substance-ontology; I would claim that such an ontology can only be logically consistent if it maintains a (moderate) nonrealistic or epiphenomenal account of temporal relations. But this would lead us too far afield. In this paper, suffice it to consider one final issue: How can we apply this ontological position on time to infinity, God, and God’s characteristic infinity?
4 The outlook Classically, we think of “infinity” in two different ways: either as temporal infinity, or as non-temporal or timeless infinity. If we consider time to be an epiphenomenon constituted by events, temporal infinity can be construed as implying an infinite or endless chain of events, just as the ancients thought that circular movements (of planets) would be infinite. Some modern cosmologists take up this ancient intuition too, applying it in the theory of an endless cosmic loop. But the validity of such a model is outside the scope of this paper. If we want to include temporal infinity in the way we understand the world, then we need this sort of endless loop consisting of endless chains of events. If we are skeptical about such loops, then we must also be skeptical about the temporal infinity. I leave aside the question of the contrary: finite timeless universes, i.e., universes in which, accidentally, no events occur; instead I will focus on another classical concept of infinity: non-temporal, timeless, or eternal infinity. This is the notion we will need to characterize divine infinity.
Temporal Relations as Epiphenomena |
49
If time is an epiphenomenon based on events, then a subject which is essentially not involved in any event would in general be non-temporal (alternative expressions include “timelessly infinite” or “eternal”). If a subject were a perfect being (otherwise known as an actus purus or an absolute simple being¹⁷), then that subject could not possibly be involved in any event. The reason is that, since all events are either beginnings, endings, or changes, involvement in one of them would by definition preclude the subject’s perfection or its ability to be simple or an actus purus or being simple. Granted that God is a perfect being, an actus purus, and simple, he must not be involved in any event, which would mean he would have to be timeless infinite or eternal. It is worth adding that, if time is epiphenomenal, timelessness is no flaw. As an epiphenomenon, time would have no being. Just as the lack of a privation would be no deficiency in being, neither is the lack of something that has no being in and of itself. God’s being can be regarded as timelessly infinite, i.e., eternal. In addition to this claim, our ontological theory of time yields other implications for thinking about God. In particular, we can fruitfully use it to discuss the puzzle of God’s relations to his creation, which indeed is temporal. Examples of such relations include his knowing temporal things and (especially) his acting in a temporal cosmos. In the following I will draw (liberally) on the interpretation put forward in William Hasker’s God, Time, and Knowledge (especially chapter 8), where Hasker argues for the principal intelligibility of God’s timeless knowing and acting in a temporal world.¹⁸ I would like to add my ontological interpretations to Hasker’s argument. The least problematic application seems to be God’s knowing, since knowing is “not a time-consuming activity” (ibid., p. 152). As I would phrase it, knowing is neither necessarily nor definitionally an event-dependent occurrence. Not every knowing is the result of a process in time; indeed, such a result would presuppose a kind of “learning-process” which, ontologically speaking, amounts to an imperfection. On the contrary: “this is a limitation of our finitude, which obviously does not apply to God” (ibid.). In other words: That a knower x knows y does not contradict the claim that x is timeless, even if y is tensed. Acting is ontologically more significant but also more problematic. Let us suppose (with Hasker) that the paradigmatic divine action in our world is God’s “preservation of the world in being from moment to moment” (ibid., p. 152). How
17 To the close conceptual connection between simplicity and immutability see Hasker (1989), pp. 182f. 18 That Hasker finally takes a critical view on God’s timelessness, especially with regard to the problem of God’s foreknowledge and free will (see ibid., chapter 9) can be neglected here.
50 | Christian Kanzian then can we understand the preservation of temporal beings by a timeless preserver? Hasker argues that such a procedure is intelligible. To do that he first discusses the intelligibility of a non-spatial God who is able – from outside of space – to bring about effects in space; Hasker concludes: “Just as the non-spatial God can act outside of space so as to produce effects at every point in space, so the timeless God can act outside of time, that is, in eternity, so as to produce effects at every point in time.” (Ibid, p. 154) The relevant distinction is between the act itself, which in the case of the timeless God is itself timeless, and the effects of the act, which may be temporal: “... the temporal characteristics of the effects of divine actions need not characterize the actions themselves. The act of preserving a temporally extended universe need not itself be temporally extended.” (Ibid., p. 158) Our sketch of the epiphenomenality of time helps us explain why it is consistent to maintain that God timelessly preserves temporally extended entities (such as things and their modes, states, and events): That these effects of God’s preservation are temporal does not affect God’s act. The reason is that the temporality of things, modes, states, and events results from the ontological function of events, which is to constitute the epiphenomenon of time.¹⁹ God preserves events, but he need not and indeed cannot preserve the epiphenomena that are constituted by these events. The same may hold for things and space: God preserves things, but not the epiphenomena that they constitute. This is no deficiency of God’s, because epiphenomena are not entities – they have no being at all. There is no deficiency in not creating or preserving non-entities. In summary, this paper aimed to develop a three-fold theory of temporality: (i) Events are the immediate constituents of temporal relations. This is what it means to say that they are four-dimensional. (ii) Temporal relations are interpreted in a moderately anti-realistic way, namely as epiphenomena. (iii) What links (i) and (ii) is the assumption that constitution is a formal internal relation (which is irreflexive, asymmetric, non-transitive); the further theoretical significance of this link, especially for a substance-ontology, cannot be pursued here. Finally, (i)-(iii) are compatible with (traditional) interpretations of infinity and of God’s eternality.
19 How the “temporal shapes” of things, modes, and states depend on events and their function to constitute temporal relation, is another ontological question which must be left for another occasion.
Temporal Relations as Epiphenomena |
Bibliography Rudder-Baker, L. (2007), The Metaphysics of Everyday Life, Cambridge: Cambridge University Press. Hasker, W. (1989), God, Time, and Knowledge, Ithaca/New York: Cornell University Press. Heil, J. (2003), From an Ontological Point of View, Oxford: Clarendon Press. Lewis, D. (1986), On the Plurality of Worlds, Oxford: Blackwell. Lowe, E. J. (1998), The Possibility of Metaphysics, Oxford: Clarendon Press. Lowe, E. J. (2006), The Four-Categorial Ontology, Oxford: Clarendon Press. Mulligan, K. (1998), “Relations – Through Thick and Thin”, in Erkenntnis: 48 (Extra-Volume: Analytical Ontology, edited by. Ch. Kanzian und E. Runggaldier), 325–353. Szatkowski, M. (ed.) (2015), God, Truth, and other Enigmas, Philosophical Analysis 65. Berlin/Munich/Boston: De Gruyter, 183–196.
51
Tomasz Kąkol
In Defense of Presentism and an Extratemporal God I begin with some definitions. “Presentism”, defined here as the claim that a time flow objectively exists along with an objective difference between past, present and future, is the stance defended by a minority of contemporary ontologists of time. Another prominent view, eternalism, claims that both the difference between past, present and future and the flow of time are merely subjective. An extratemporal God (well-known in traditional rationalist metaphysics¹) is often thought of as the opposite of “the living God”: philosophers since Aristotle (not to mention common sense) have held that change implies time and therefore, extratemporality entails unchangeability. One can argue that my definition of “presentism” is idiosyncratic; specifically, one usually understands this doctrine as stating that only the present (but neither the past nor the future) exists. This is true,² but I deliberately depart from this unfortunate custom since it creates an ontological straw man that is too easily refuted.³ In my paper I defend presentism using two positive and two negative arguments: 1) the presentist “now” is assumed in contemporary physics, arguments to the contrary notwithstanding; 2) presentism has explanatory power, and can explain many phenomena such as the permanent change in our temporal perspective and our concern about the future and the asymmetry of biological growth⁴; 3) the putative explanation of the intuition of time flow is ungrounded, and the same applies to both the entropic theory of time and the causal theory of time; and 4) presentism neither implies relative existence (when combined with special relativity) nor is subject to the notorious “how fast does time flow” objection and the alleged rejection of the so-called “truthmaker principle”.
1 One of the main reasons for holding that God is extratemporal is that a temporal God cannot both be omniscient and respect human freedom. If humans act freely, then the outcomes of their deeds cannot be predicted with 100% certainty, whereas an extratemporal being need not predict anything. 2 See, e.g., Sider (2001), Hansson Wahlberg (2009) or Grygianiec (2011) 3 See, e.g., Dummett (2004), p. 74 – if presentism (as “usually” understood!) is true, then the present “is a mere boundary between past and future. But a boundary can exist only if that which it bounds exists.” So, nothing would exist! 4 Such as the change “from the kernel to the mature tree”. See Gołosz (2011), p. 152. https://doi.org/9783110594164-004
54 | Tomasz Kąkol Although I owe those arguments to Jerzy Gołosz (the most vigorous proponent of presentism in contemporary Polish philosophy), the ontology of time that I propose is more Ingardenian in spirit and more moderate: in particular, I disagree with Gołosz’s theses that (a) presentism entails endurantism⁵ and (b) the famous solutions of Einstein’s field equations of gravitation with closed time-like curves do not entail that presentism only contingently applies to our world.⁶ As regards God, we have a mathematical model of tenseless dynamics (Michał Heller) that not only answers the objection that I mentioned but also makes sense of the traditional “dark” conception of creatio continua. * Ad 1. A. Einstein famously wrote to the widow of his friend that “the [objective] distinction between past, present and future is only an illusion, even if a stubborn one”⁷. But this remark raises a question: why do physicists stubbornly talk – as, for example, Stephen Hawking does in his A Brief History of Time – about the present rate of the expansion of the universe or the present value of the density of matter⁸ or the present value of background radiation⁹ or the present bond between electric and magnetic fields¹⁰ and so on. Einstein is suggesting to the widow that eternalism is true, but if it is, why do physicists often talk as if presentism is true. In short, presentist language is ubiquitous in contemporary physics. Ad 2. As time flows, we undergo a permanent change in our temporal perspective; some (Einstein, Hawking in the book mentioned¹¹ and, interestingly, even poets such as R. M. Rilke in his Letters to a Young Poet¹²) try to explain this by claiming that moving bodies such as the Earth in fact move along a straight path in four-dimensional spacetime. However, this is inconsistent with the fourdimensional, eternalist picture of a static or block universe. To put it another way, if we talk about moving in a n-dimensional structure, this raises the objection that we thereby are just assuming an objective time. Moreover, eternalists need to explain why we have more interest in the future than the past if eternalism is true. One attempt, by Paul Horwich, grounds this
5 Conf. idem, p. 9. 6 Conf. idem, p. 133. 7 The quote is from Davies (1995), pp. 70, 76. 8 See, Hawking (1998), p. 27. 9 See idem, p. 98. 10 See idem, p. 98. 11 See idem, pp. 20 and 68. 12 Conf. Rilke (1977), p. 70.
In Defense of Presentism and an Extratemporal God | 55
concern in biological fitness, but the truth is that this is so only because there is asymmetry between the past and the future (the past is unchangeable), which calls for further explanation.¹³ Ad 3. The usual explanation of the alleged illusion of time flow is that our cognitive processes purportedly produce this illusion. But the problem is that they are, simply as experienced processes, temporal and dynamic entities.¹⁴ As for probably the most popular theory of time, namely the entropic theory of time, if time is identified with or at least correlated with (or determined by) the growth of entropy, then what is happening in regions in which reversible processes hold since the statistical character of the second law of thermodynamics allows for such realities? Is time static there? Or does the flow of time also reverse? The former can reasonably be rejected. Some people say that the problem is that, viewed in this manner, the second law of thermodynamics becomes trivial.¹⁵ I think otherwise, since the fact that water is H2 O is not trivial. A much more serious difficulty is why we remember the past and not the future if we do not concede, following Hawking, that if entropy were decreasing, then we would remember the future!¹⁶ Indeed, his firm conviction unfortunately looks like fantasy. The causal theory of time, another prominent view, also faces objections. First, as was observed long ago, causality at most presupposes time, but it is not implied by it¹⁷ (at most, since we should not exclude a priori non-temporal causality). Second, if causality is guaranteed by the laws of physics, which concern (nowadays four distinct) fundamental forces (strong nuclear, weak nuclear, gravitational and electromagnetic), these forces – apart from the second one which can be ignored from the macroscopic point of view – are insensitive to the direction of the passage of time, so to speak.¹⁸
13 See Gołosz (2011), p. 31. 14 See idem, pp. 24, 26, 32. Some readers may accuse me of contradiction since I mentioned tenseless dynamics and later I will talk about the “Gödelian world” which is in motion but without time flow. I reply that the model of tenseless dynamics comes from microphysics and should not be extrapolated to the world of such “medium-sized phenomena” as our mental states and processes (it is consistent with the utility of this model in showing that an extratemporal God can be “dynamic” since we do not have any experiential access to Him/Her, whereas we do have access, at least partly, to our mental states – through introspection). Similar points should be made about “Gödelian world” as we will see in a while. 15 See Gołosz (2011), pp. 162f. 16 See Hawking (1998), pp. 73 and 75. 17 See Gołosz (2011), pp. 162, 171. 18 See idem, p. 176.
56 | Tomasz Kąkol Ad 4. Presentism does not entail relative existence when combined with special relativity; presentness is just relativized to the frame of reference (recall that the “relativity” in special relativity does not imply non-objectivity). As for the objection that presentism contradicts special relativity because different objects in different frames of reference can be regarded as simultaneous or co-present or coexisting (in the tense sense of the term)¹⁹ – it should not be a reason for worry since the causal structure of spacetime is saved across the frames of reference.²⁰ Similarly, the alleged problem of the truthmaker is straightforwardly solved: the truthmaker of, say, starting World War II does not occur (or, to stress the tense sense of the term, is not occurring now), since it has already ocurred.²¹ To answer the question “how fast does time flow?” we need to introduce more ontology. Following E. Husserl, R. Ingarden makes a distinction between pure and empirical possibilities (and other modalities), restricting ontological considerations to the former only. Closing the first volume of his monumental “Controversy over the Existence of the World” (I mean here the last, complete edition, i.e. from 1987), Ingarden gives the characteristics of eight possible modes of real existence (as opposed to the absolute, the ideal and the purely intentional). Ingarden divides these eight modes into three groups called the presence, the past and the future. In other words, time in general (in specie) is for Ingarden the mode of existence of real objects (or, equivalently, the mode of real existence or the real mode of existence – according to Ingarden, all these expressions have the same meaning, although he prefers the phrase “the real mode of existence”). Remembering that Ingarden states that every object is a triunity of matter, form and the mode of existence (matter and form understood in a non-Aristotelian way), he follows the tradition according to which there is a very close link (to say the least) between the concept of time and the concept of existence. To highlight this connection (as it is found in the ambiguity of the English word “presence” and in many other languages – but, interestingly, not for example in Polish) I propose to name – in a Heideggerian manner – the past wasence and the future willbence. Next, Ingarden divides real beings into three classes: objects enduring through time (substances, using traditional terms, or endureres), processes, and events. They have different forms and, consequently, different modes of existence: enduring through time or short endurance, becoming and (unique) occurring or occurrence respectively (unique, since Ingarden is convinced that events never recur or return). In effect, we have the following table of time:
19 See Hansson Wahlberg (2009), pp. 21–23. 20 See Gołosz (2011), p. 127. 21 Conf. idem, p. 98.
In Defense of Presentism and an Extratemporal God | 57
In this way – contrary to, e.g., St Augustine – we can clearly tell what time is: first, time is always the time of something. For example, the past of a pen is its past endurance, the presence of this process of my writing is its present becoming and the future of my departure to Budapest (assuming that it is an event – some people could argue that it is more of a process) is its future occurrence. Secondly, the Ingardenian conception can meet the challenge of the notorious “how fast does time flow” charge. To begin with, the becoming of the process of reddening, say, of a spoon for babies (there are such spoons, made of material that change color depending on the temperature of milk or another kind of drink) can be faster or slower, but when we choose the model time – for example, one becoming of a full revolution of the pointer of a certain gadget/machine – we can measure time saying for example that the reddening process of the first spoon is twice as fast as the reddening process of the second spoon, since the former takes up half a model time, whereas the latter takes up the whole of it. If a critic were to say that we have now introduced a vicious circle or a regress, as we can ask about the velocity of the becoming of this full revolution, we could reply that if she were right we could not ask for example “how long is the length of the model?” (the late Wittgenstein thought so but he was rightly criticized by Kripke). The only inconvenience is linguistic, since it sounds bizarre that the mode of existence can be faster or slower. * Data from physics suggest that every presentist ontology of time, Ingarden’s included, should be limited for at least two reasons. First, in 1924 Cornelius Lanczos, later Einstein’s assistant, discovered some very peculiar solutions of Einstein’s field equations of gravitation. They were in a sense rediscovered by Willem Jacob von Stockum in 1937, and then by Kurt Gödel in 1949, and, due to the latter’s fa-
58 | Tomasz Kąkol mous name, they are often referred to as “Gödelian solutions”²². In general relativity we treat space and time inseparably, as spacetime. In this way we can represent within this structure histories of objects with non-zero rest masses (such as atoms or electrons) as certain curves (physicists call them “timelike curves”). Imagine a caricature of a Heraclitan (or Stoic or Nietzschean) world of the so-called eternal return that would consist of two spatial dimensions only and two small point masses rotating around each other. When we add a temporal dimension, we obtain two timelike curves similar to a double helix (very well-known from biology). However, the peculiarity of Gödelian solutions is that there are closed timelike curves there that entail the prima facie possibility of time travel. It should be stressed here that Gödel’s universe cannot model ours, since, apart from its global rotation, it is static, whereas ours is expanding (moreover, Gödel’s world is simplified, as its spacetime is homogeneously occupied by dust with constant density); nevertheless, it shows that presentism is probably contingently only true of our world. The reason is simple: closed time excludes both the time flow and the division into past, present and future. Imagine that we choose a point on a closed timelike curve as “now”. According to presentism, what was before, there is no more, but before was the very same point, which amounts to a contradiction. As for time travel, it does not make sense to interpret it as going back to the past, since this is not “the past” in the presentist sense. When someone thinks that there is a trick here in that in this way we have only shown that the Gödelian universe is inconsistent, recall that something similar – and equally paradoxical – holds in special relativity concerning photons (that is, quanta of electromagnetic force): according to special relativity, there is no lapse of time in them. Nevertheless, photons are constantly in movement, indeed the fastest physical movement! Consequently, presentism in all probability is both contingent and local. Its locality is additionally suggested by several cosmological models of (semi)quantum gravity. According to one of them, that is, the well-known (and criticized) model of Jim Hartle and Stephen Hawking from 1983, there is no time when we cross the famous Planck’s thresholds (Planck’s time ≈ 10−44 s, Planck’s length ≈ 10−33 cm, Planck’s density ≈ 1093 g/cm3 ). The much more interesting one is the more radical: there is neither time nor space when we cross these magic boundaries; this view has been popularized in Michael Heller’s book under the very meaningful title The Beginning is Everywhere (“and at anytime” should be added) in 2002 (in Polish).²³ This model not only makes “desingularization of space-time” successful (as Hawking’s
22 See Heller (2012), pp. 48–50. 23 See Heller (2002).
In Defense of Presentism and an Extratemporal God |
59
does)²⁴ but also solves the so-called horizon problem without ad hoc inflation (almost every contemporary cosmological model assumes mysterious inflation – that is, the rapid expansion of the universe) and, in addition, explains the notorious Einstein-Podolsky-Rosen paradox of non-locality.²⁵ Setting aside mathematical details, Heller’s model uses non-commutative, pointless space. Since (“standard”) space and time imply locality, because they consist of points (interpreted as moments when time is concerned) or extended simples, as some – Ingarden included – think, which are local entities, non-locality excludes both space and time. The consequences of this approach, if Heller is right, are philosophically far reaching: it not only denies naïve materialism stating, as David Armstrong would put it, that the totality of being is no more than the spatiotemporal system, and denies the phenomenologically and commonsensically obvious proposition that dynamics or change implies time, but also entails that Ingarden’s conviction that time is the mode of existence of all real objects is unsupported. The consistency of tenseless dynamics is obtained by defining in the space that Heller uses analogs of (force) vector fields²⁶, whereas the similarity to the traditional doctrine of creatio continua is obvious: when we arbitrarily choose at time t any physical thing A in our vicinity (say, the table I am sitting at) and we are mentally (i.e., using our imagination) approaching Planck’s length, we are approaching Planck’s density and the region (or “the region”), where there is neither space nor time – and this is just the region sought by cosmologists debating about the so-called Big Bang and its vicinity. Thus, the “true” beginning, or the ultimate origin of my table, is borne out at t. Of course, this model also poses questions traditionally directed to the theory of creatio continua, such as whether we have in effect overdetermination (since prima facie my table would be both caused by the carpenter somewhere in the past, and yet constantly borne out), but this is a topic for another article.
Bibliography Davies, P. (1995), About Time: Einstein’s Unfinished Revolution, London: Penguin Books. Dummett, M. (2004), Truth and the Past, New York: Columbia University Press.
24 The so-called singularity is known even among laymen just as black holes are. 25 Concerning those problems see Heller, idem, chs. 10f. Notice especially that the horizon problem is interpreted as also indicating non-locality. 26 See idem, ch. 9.
60 | Tomasz Kąkol Gołosz, J. (2011), Upływ czasu i ontologia [The Flow of Time and Ontology], Kraków: Wydawnictwo Uniwersytetu Jagielońskiego. Grygianiec, M. (2011), “Trwanie w czasie” [“Persistence in Time”], in Przewodnik po metafizyce [A Guide to Metaphysics], edited by S. T. Kołodziejczyk, Kraków: WAM, 211–276. Hawking, S. (1998), A Brief History of Time, New York/London/Toronto/Sydney/Auckland: Bantam Books. Text available at: https://www.pdf-archive.com/2014/07/24/a-brief-historyof-time/ (access: 1.09.2016). Rilke, R. M. (1977), Lettres à un jeune poète, Paris. Hansson Wahlberg, T. (2009), Objects in Time. Studies of Persistence in B-Time, Lund University (doctoral dissertation). Heller, M. (2012), Ostateczne wyjaśnienia wszechświata [The Ultimate Explanations of the Universe], Krakow: Universitas. Heller, M. (2002), Początek jest wszędzie [The Beginning is Everywhere], Warszawa: Prószyński i S-ka. In English this model was presented in the article “The Emergence of Time”, in Physics Letters A, 250(1-3), 48–54. Sider, T. (2001), Four-dimensionalism. An Ontology of Persistence and Time, Oxford: Clarendon Press.
Srećko Kovač
Concepts, Space-and-Time, Metaphysics (Kant and the dialogue of John 4) 1 Introduction In an empirically oriented common-sense ontology, first-order concepts are expected to be “concrete” and to denote sensible objects given in space and time, while other, “abstract”, concepts should denote words, sentences, sets, numbers, or concepts themselves, possibly of a questionable ontological status, or, moreover, conceived merely as a manner of speaking, subjective representations, or “ideas” without an actually corresponding reality. In a formalized presentation, such an empirical theory would have a model comprising a first-order domain of sensible objects denoted (possibly in n-tuples) by predicates. The domain itself and the relations on the domain, as well as syntactic “objects” (terms, predicates, formulas – replacing concepts and judgments), remain abstract, metatheoretical entities that are not empirically given for the object theory. Besides, the domain and the relations on the domain may appear as members of a second-order domain if the formalization is extended to a higher-order setting, but, of course, this still does not make the first-order domain and the relations on it themselves empirically existing objects. Some essential features of abstract, model-theoretic, concepts of a possible empirical theory are traceable back to Immanuel Kant’s “transcendental logic” (with some characteristic differences).¹ Against this background, we examine the objective reality of the abstract concepts involved, putting them in the context of a possible religious experience as presented in the text of John 4.
2 From metatheory to metaphysical theory It can be recognized that Kant’s theory of transcendental ideas serves as a sort of first-order model for empirical reasoning and knowledge, where transcendental ideas represent three sorts of totalities of conditions of empirical knowledge:
1 For some significant connections of Kant’s logical theory with modern logic, see, e.g., Achourioti and van Lambalgen (2011) and Tiles (2004). https://doi.org/9783110594164-005
62 | Srećko Kovač (a) the totality with respect to a subject (“complete subject”,² never occurring as a predicate, B 379): “I” (“mere consciousness”, “determining Self”), which thinks, is the metatheoretical subject “X” of all thoughts (e.g., of concepts), which are its predicates (B 404, A 402);³ (b) the totality of the “series” of conditions (“world”) of an empirically given object: each such object is possible only if the whole series of its conditions, too, is in some way already given (B 436);⁴ (c) the totality of concepts as predicates (“the sum total of all predicates”) – as if comprised in some common “ground” (B 607): “the most real being” (ens realissimum).⁵
These “transcendental” ideas do not belong to empirical knowledge as an objecttheory, but to its metatheory. Kant further specifies this by assigning those ideas a non-constitutive, regulative (and heuristic, B 644) role for empirical knowledge.⁶ We will now focus on some structural similarities between Kant’s system of transcendental ideas and the conceptual structure of religious knowledge (religious belief)⁷ as presented in Jesus’ dialogue with a Samaritan woman in John 4.⁸ From the standpoint of religious knowledge, transcendental (metalogical) concepts obtain their specific objective reality and become metaphysical concepts
2 Kant (1910–), Vol. IV, p. 330. 3 “I” is not a concept (or any representation) of an object, but just a general “form” of the knowledge of an object, since, as Kant points out, only by means of it do “I think anything” (B 404; cf. “I” is “that which I must presuppose in order to cognize any object”, A 402). In this sense, we find “I” replaced in a standard first-order model simply by a chosen set of objects (domain). According to Kant’s theory, the application of “I think” is restricted to the “manifold” (Mannigfaltigkeit) of what is given in a sensible spatio-temporal intuition. Let us note that the unity of a concept in Kant’s (intensional) theory originates from the “analytical unity of consciousness”, which “pertains to” the concept, while the unity of a concept (predicate) in a standard (extensional) first-order theory model-theoretically derives from the set itself (a subset of the domain) that is assigned to the concept as its extension. 4 This is in accordance with the “principle of reason”: “if the conditioned is given, the whole sum of conditions, and hence the absolutely unconditioned, is also given, through which alone the conditioned was possible)” (B 436). In standard first-order model theory, element (b) of Kant’s model is replaced by a relational structure that is imposed on the domain by the interpretation of relation symbols and complemented by the conditions of the satisfaction of formulas. 5 Cf. Kant’s “principle of thoroughgoing determination”, according to which each object should be determined with respect to each concept (B 599–600). In a first-order theory, the interpretation of one-place relation symbols replaces element (c). 6 For Kant, transcendental ideas are “regulative principles for the systematic unity of the manifold of empirical cognition in general” (B 699). 7 See on religious belief and knowledge in the introductory chapter of Kovač (2015a). 8 For a theological interpretations of this episode, see, for instance, Jojko (2012), Botha (1991) and Varghese (2009); for a theological-historical context, see, e.g., Novakovic (2013). Here, we further elaborate our initial analysis in Kovač (2011).
Concepts, Space-and-Time, Metaphysics |
63
(soul, world, God); in addition, the application of concepts in general extends to the realm of non-sensible objects (“noumena”).⁹ We first summarize the progress of Jesus’ dialogue with the Samaritan woman (with a slight rephrasing) in Figures 1 and 2. J: Give me a drink.
J: If you knew the gift of God, and who is S: You (a Jew) cannot ask me (a Samaritan speaking with you, you would have asked him, and he would have given you living water. woman) for a drink. × S: You do not J: The water I shall give will become a spring of water even have welling up to eternal life. a bucket. × S: Sir, give me this water.
[see Figure 2]
Fig. 1
We now briefly informally analyze the dialogue in John 4, comparing it with Kant’s system of transcendental (metatheoretical) ideas.
9 Cf. B 395, footnote, with Kant’s following critical remark: “Metaphysics has as the proper end of its investigation only three ideas: God, freedom, and immortality [. . . ] The insight into these ideas would make theology, morals, and, through their combination, religion, thus the highest ends of our existence, dependent solely on the faculty of speculative reason and on nothing else”. According to Kant, in the process of work we should proceed “from what experience makes immediately available to us, from the doctrine of the soul, to the doctrine of the world, and from there all the way to the cognition of God”.
64 | Srećko Kovač [continuation from Figure 1, S: Sir, give me this water ]
J: Go call your husband.
S: I do not have a husband. ×
J: You are right. You have had five husbands, and the present one is not your husband.
S: You are a prophet. Our ancestors worshiped on Gerizim, you people worship in Jerusalem. ×
J: Now the worship must be in Spirit and truth. S: The Messiah J: I am. will tell us everything. ×
Fig. 2
2.1 Self The idea of “self” (“I”, “you”, implicit in “we”), with knowledge, belief, speaking, and being as belonging to it, explicitly occurs and has an essential role in the dialogue of John 4. Two agents (selves), Jesus and a Samaritan woman, are engaged in the dialogue that is advancing step by step through logical reasoning and through a gradual evolving of new knowledge. The logical idea of “I think” is recognizable where “I” is mentioned as a subject of epistemic (mental and verbal) acts.¹⁰ In particular, the logical aspect of “self” is recognizable in the fact that the dialogue proceeds through the considering of and finding solutions for the contradictions appearing during the conversation. As will be seen, contradictions, as a means of a possible questioning, or
10 Cf., for example, “if you knew the gift of God and who is saying to you [. . . ]”, John 4:10; “you are right in saying [. . . ]”, John 4:17; “what you have said is true”, John 4:18; “I can see [the¯or¯o] that you are a prophet”, John 4:19; “we worship what we understand”, John 4:22; “I know that the Messiah is coming”, John 4:25.
Concepts, Space-and-Time, Metaphysics |
65
at least of a clarification, of the theses that appear in the dialogue, are the main logical vehicles of the conversation between Jesus and the Samaritan woman (see the next subsection). In the Gospel, “self” is not just an abstract, regulative idea, but denotes a being, at first only an empirical being (a Jew, a Samaritan woman; John 4:9),¹¹ and eventually a being “in Spirit and truth”.¹² According to Kant, in metaphysical psychology a paralogism occurs that is based on the non-justified assumption of an intuitive givenness of some persistent “self”, which leads to the “inference” that “self” is a substance (B 411).¹³ In distinction, the Gospel proposes a justification of the existence of “self” as a permanently (“eternally”) given subject of religious (non-sensible, spiritual) experience.¹⁴
11 Cf. Kant’s “empirical unity of consciousness”: “One person combines the representation of a certain word with one thing, another with something else; and the unity of consciousness in that which is empirical is not, with regard to that which is given, necessarily and universally valid” (B 140). 12 It does not suffice for a worshiping agent to be in space and time, since the worship should take place “in Spirit and truth”: “God is Spirit, and those who worship him must worship in Spirit and truth” (John 4:24). The spiritual existence of “self” in truth is explicit in Jesus’ enunciation of his own being: “I am [the Messiah], the one who is speaking with you” (John 4:26; cf. subsection 2.3 below). 13 “Thus if that concept, by means of the term ‘substance’, is to indicate an object that can be given, [. . . ] then it must be grounded on a persisting intuition as the indispensable condition [. . . ] through which alone an object is given [. . . ]. But now we have in inner intuition nothing at all that persists, for the ‘I’ is only the consciousness of my thinking.” (B 412–413). 14 Even in Kant’s moral philosophy, “self” is merely a postulate, not a concept that denotes objective reality. If we want to avoid the paralogism of metaphysical psychology (Kant, B 411) by means of a new sort of knowledge proposed in John 4, we get the following correct syllogism: What cannot be thought otherwise than as subject does not exist otherwise than as subject, and is therefore substance. A thinking being, considered as existing in Spirit and truth, cannot be thought otherwise than as subject. ⊢ A thinking being, considered as existing in Spirit and truth, exists only as subject, i.e., as substance. The expression “as existing in Spirit and truth” replaces Kant’s “considered merely as such”. In this way, “thinking” (“be thought”) in the middle term is in the minor premise, too, understood with respect to a given (existing) object, not just with respect to the subject of thought (B 411, note).
66 | Srećko Kovač
2.2 World We now proceed to a comparison of the dialogue of John 4 with Kant’s transcendental idea of world – the totality of a series of conditions of empirically given objects and states (B 391) – and with Kant’s corresponding cosmological antinomies. We will see that Kant’s four conceptual aspects of a possible world totality – regarding (a) the composition of the whole of space-and-time, (b) the divisibility of matter, (c) causality, and (d) dependence in existence (B 438–443) – can also be found in John 4 (in a somewhat different way), as emerging one after another in the contradictions and their solutions unfolding during the dialogue.
Antinomies Kant’s antinomies arise from the question whether a given object and its state have some first condition (be it immediately given or not) or whether the series of its conditions is infinite (B 445–446). However, in John 4 it seems to be assumed that the world has the beginning in all four aspects (a) – (d) mentioned above (the beginning of time and space, the origin of matter, the first cause, and the unconditioned existence). So, the antinomies in John 4 arise from the problem whether the first condition (beginning) is immediately given (present) or whether it is given only through a (possibly long) series of intermediate conditions.
(a) Extension of space and time Jesus: Give me a drink. Samaritan woman: How can you, a Jew, ask me, a Samaritan woman, for a drink? (For Jews use nothing in common with Samaritans.) ⊢ Samaritan woman: [Contradiction]. (Cf. John 4:7–10).
Is a drink immediately available to Jesus, or should he look for it elsewhere and from someone else? To a significant extent, this can be interpreted in terms of Kant’s first cosmological antinomy (B 454, 455), taking a drink as representing life (cf. “living water”, John 4:10), and, in connection with this, as symbolizing time and space (flow and places, i.e., history, of life): do I find the beginning of my life, i.e., of my time and space, immediately here and now (thesis), or should I return to the origins of my (Jewish) past, and go out of this (Samarian) place back to the land of my origin (homeland, Galilee) (antithesis)?
Concepts, Space-and-Time, Metaphysics |
67
Thesis 1: The beginning of an agent’s a time and the beginning of a’s space are always immediately present. Antithesis 2: The beginning of an agent’s a time lies in the (past) moment of the beginning of a’s life, and the beginning of a’s space lies in the (distant) place of a’s origin.
Instead of Kant’s opposition between the finite and the infinite regress in space and time, here we encounter the opposition between the immediate presence of the beginning of time and space, and its indirect givenness by means of intermediate segments of time and space.¹⁵ Another interesting distinction appears regarding the conception of the beginning of space: in Kant, it is conceived as a possible end (outer limit) of space, while in John 4 it seems to be understood as the origin of space in relation to “self” (i.e., as home, homeland).
(b) Divisibility of matter Jesus: If you knew the gift of God and x such that Jesus=x, you would have asked x and x would have given you living water. Samaritan woman: Sir, you do not even have a bucket and the cistern is deep. ⊢ Samaritan woman: [Contradiction, except that Jesus is greater than Jacob]. (Cf. John 4:10–11).
Can water be reached from the deep well and given without a bucket? In other words, to come closer to the terms of Kant’s second cosmological antinomy (B 440, 443), can matter (reality) be immediately given, without any partitioning (thesis), or is matter given only piecemeal, in portions consisting of some elementary units (“a bucket”) (antithesis)? Thesis 2: Matter can be immediately given to an agent a without any partition of matter. Antithesis 2: Matter is given to an agent a only in portions consisting of units.
Again, we note a difference: in Kant’s antinomy, there is the opposition between the finite and the infinite partitioning of matter, whereas in John 4 the opposition is between the immediate givenness of matter, and the givenness of matter only by means of its partition (possibly as a long series of portions of matter).¹⁶ Besides,
15 The problem of mediation is already announced in the introduction to John 4: “[. . . ] he [Jesus] left Judea and returned to Galilee. He had to pass through Samaria” (John 4:3–4). 16 Cf. for instance, “Sir, give me this water, so that I may not be thirsty or have to keep coming here to draw water” (John 4:15). That is, it is assumed that water is usually available only as a
68 | Srećko Kovač we remark that Kant conceived the divisibility of matter by assuming that a thing to be divided (possibly into simple parts) is already given. In John 4, in distinction, the question regarding matter (“water”) is whether it is available at all prior to its partition (thus, it is not atomism which is the central problem).
(c) Causation Samaritan woman: Sir, give me this water [i.e., living water, which will become a spring of water welling up to eternal life]. Jesus: Go call your husband and come back. Samaritan woman: I do not have a husband. Jesus: You have had five husbands, and the one you have now is not your husband. What you have said is true. ⊢ Samaritan woman: [Contradiction]. (Cf. John 4:14–18).
Has the Samaritan woman an immediately present first ground of her own wish to get living water – thesis; or is her wish grounded on the whole causal series of states and events going back from the present to the past times out of her reach (five past husbands, with the present “non-husband”) – antithesis? We can recognize an analogy with Kant’s (third) antinomy of the causality of freedom and an endless series of the preceding causal events (B 441–442, 443), modified here into the antinomy between the possibility that someone freely determines her (his) own wish/will (thesis; cf. the Samaritan woman’s wish as a possible free beginning of a new causal series), and the determination by a (long) series of the preceding states (antithesis; the Samaritan woman’s determination by her past states). Thesis 3: An agent a can freely begin a causal series by means of a’s wish or will. Antithesis 3: An agent a is determined by the preceding causal series of states.
It is interesting to note that in this antinomy the concept of causal beginning, both in John 4 and Kant, is related to free will/wish. However, in John 4 the (Samaritan woman’s) wish is, at first, not directly related to an action, but to someone else’s (Jesus’) will (“Sir, give me this water”). We will come later (solutions below) to a related crucial difference from Kant’s conception.
(long) series of portions of water drawn from the well.
Concepts, Space-and-Time, Metaphysics |
69
(d) Dependence in existence Samaritan woman: Sir, I can see that you are a prophet. Our ancestors worshiped on this mountain [Gerizim], but you people say that the place to worship is in Jerusalem. ⊢ [Contradiction]. (Cf. John 4:19–20).
Are the place and time of the presence of God (in worship) necessary in itself, i.e., independent of any further condition – thesis; or are the place and time of worship contingent (e.g., Mount Gerizim for the Samaritans, Jerusalem for the Jews), i.e., dependent, for example, on long sequences of different traditions¹⁷ – antithesis? The thesis is not explicit in the text, but is implicitly contained in the antithesis, which is in itself contradictory (the traditions mentioned are mutually exclusive),¹⁸ as well as in the solution (see below). Thesis 4: God can be immediately present to an agent a independently of any contingent place and time. Antithesis 4: God is present to an agent a only in dependence of some contingent place and time.
We can compare this antinomy of John 4 with Kant’s antinomy between the thesis that there is an absolutely necessarily existing being (the world itself, or some being outside the world), and the antithesis, according to which all beings exist contingently (cf. B 442, 443). In the context of John 4, the existence of God is already presupposed in the solution of the third antinomy (see solution (c) below). Hence, the fourth antinomy specifically concerns the question of the existence of God; in addition, it concerns His presence – not just some necessary existence, possibly completely separated from the events in the world.
17 As concluded in Novakovic (2013), p. 215, “the main difference between the Jews and the Samaritans was not their ethnicity or religiosity but the location of their cultic center”. 18 “There is no doubt that the building of the Gerizim temple [. . . ] was met with disapproval by the Jews.” As to the Samaritans, “Mount Gerizim [. . . ] continued to be regarded as the only legitimate place of worship”. (Novakovic (2013), pp. 212–213).
70 | Srećko Kovač Solutions Solutions of the antinomies in John 4 differ from Kantian solutions in that they solve each antinomy by affirming the immediate presence in Spirit of the (unconditioned) totality (thesis), while confining the antithesis (the totality accessible only through a possibly long series of conditions) merely to the sensible empirical world. Only thesis is knowledge (“seeing”) and about the true reality, while antithesis is ignorance.¹⁹ In this way, a duplication of concepts arises (explicitly or implicitly): physical water and spiritual water, physical well and spiritual well (spring), physical giving/receiving and spiritual giving/receiving, physical drink and spiritual drink, physical thirst and spiritual thirst, time and space of the physical world and “time and space” of “Spirit and truth”, physical husband and true husband, the truth of the physical world and spiritual truth, father (e.g., “our father Jacob”, John 4:12) and Father, worship in a physical world and spiritual worship, physical self and spiritual self. By the term “physical”, we intend to cover both the “natural” and the “historical” objects and states-of-affairs.²⁰ As is well known, Kant gave a positive solution both for the thesis and for the antithesis only in the case of the third and the fourth antinomies: reducing them to subcontrarieties by distinguishing “sensible” from “intelligible” causality and dependence of existence.²¹ On the other side, he resolved the first and the second antinomy by negating the thesis as well as the antithesis: reducing them to contraries because of the impossibility of the contradictory subject: the whole of time and space cannot be given in time and space, the whole partition of a material thing cannot be given in the thing, since this thing, which is, according to Kant, only our representation, is really divided only to the point up to which it is at a moment actually divided in our representation.²² Regarding intelligible causality, let us recall that only the moral causality of freedom (connected with the third antinomy) received in Kant, in his practical philosophy, the status of objective reality.
19 For instance, “[. . . ] we know that this is truly the savior of the world” (John 4:42). “You people worship what you do not understand [ouk oidate]; we worship what we understand [oidamen]” (John 4:22). “[. . . ] no one can see the kingdom of God without being born from above” (John 3:3). 20 See Kovač (2015a) for the distinction of naturalistic and historicist conceptions of knowledge in John 3. 21 B 566, 587–589 and Prolegomena, Kant (1910–), Vol. 4, pp. 343–347, §53. 22 Cf. Kant’s clarification in Prolegomena (Kant (1910–), Vol. 4, pp. 341–342, §52c), and in B 545– 555.
Concepts, Space-and-Time, Metaphysics |
71
(a) The first antinomy (extension of time and space). The solution confirms the thesis that the beginning of time and space is immediately present (Thesis 1) – as a “gift of God”, while in the sensible world the beginning of time and space should be looked for (back) in the time and place of one’s origin (Antithesis 1). In the Gospel, Jesus gives the solution by distinguishing between sensible empirical water and “living water”, which is a “gift of God”,²³ and which he presently possesses and could give to the Samaritan woman (cf. the first premise of the second antinomy above, John 4:10, and Figure 1). We note a specific way in which Thesis 1 itself is confirmed: living water is immediately available simply by asking for it in the right way (as for a gift of God). Elaborating this a bit further, Thesis 1 leads to the equivalence relation (accessibility in time) on space points on the ground of the outer infimum (God), which is immediately related to a chosen inside point (Jesus, who comes from God; see John 4:25–26), and on the assumption of transitivity²⁴ and euclidity.²⁵ Of course, in a sensible world, there could be barriers which prevent accessibility in space and time. This solution underlies the second antinomy (see above): whether matter (the content of space and time) can be given in space and time immediately, without any partition, or only piecemeal? (b) The second antinomy (divisibility of matter). The solution affirms Thesis 2, according to which an indivisible origin of matter is immediately present, prior to any partition. This solution restricts the validity of Antithesis 2 to the sensible world, where we encounter only a (possibly long) series of portions of dividable matter, and it relates the validity of Thesis 2 to non-physical matter and “eternity”.²⁶ In the words of the Gospel, what Jesus will give is not water in parts (in
23 According to Genesis 1:1–2, “in the beginning”, before the creation of light, there was “a mighty wind sweeping over the waters” as a “pre-creation state”. See the comment on this place in (New American Bible (2011), p. 10, note *). 24 Transitivity is indicated, for example, by the above-quoted sentence “If you knew the gift of God [. . . ]” (antinomy (b), also Figure 1). Besides, see later in the text (assuming that “water” is, in a way, present in words): “Many of the Samaritans of that town began to believe in him because of the word of the woman” (John 4:39). 25 As an indication, we remark that the Samaritan woman, as well as other Samaritans in her town, heard Jesus’ words, and then the Samaritans confirmed their belief to her: “[. . . ] they said to the woman: ‘We no longer believe because of your word; for we have heard for ourselves, and we know that this is truly the savior of the world’ ” (John 4:42). 26 The possibility and the presuppositions of something like “eternal life”, as well as the questioning of the received views on it, are the subject of the third and, eventually, the fourth antinomies.
72 | Srećko Kovač buckets), from some gradually accessible distance (from Jacob’s deep well), but the spring of water and of eternal life: Everyone who drinks this water will be thirsty again; but whoever drinks the water I shall give will never thirst; [. . . ] [it] will become in him a spring of water welling up to eternal life. (John 4:13–14).
This solution leads to a new (third) antinomy: to live in dependence on the physical world and its causal sequences seems to contradict the possibility to access the origin of eternal life. The problem should be answered by the next solution. (c) The third antinomy (causation). The solution includes the affirmation of Thesis 3, about the causality of one’s own wish (the possibility of a new causal beginning) through a (liberating) relationship to truth and God, as well as the affirmation of Antithesis 3, about an agent’s dependence on a (long) causal sequence (e.g., tradition, personal history), as confined to the sensible empirical world. Thus, the Samaritan woman, living under the conditions of her past life (a sequence of husbands or “husbands”) faces this fact (truth), and converts to the worship of God (in a way, she recognizes that truth comes from God).²⁷ The conversion to truth and the worship of God should make her free and able to break the preceding causal sequence of events, and to start a new sequence.²⁸ Such a new start does not exclude, but could perhaps rather require, having a true husband.²⁹ In a still wider perspective, the whole long causal sequence (history) consisting of “fathers” and “ancestors” (e.g., Jacob, Joseph) is now being replaced with the one present cause: the Father, as God is named by Jesus (John 4:21).³⁰
27 Cf. later in the Gospel: “I tell you what I have seen in the Father’s presence”, “you are trying to kill me, a man who has told you the truth that I heard from God” (John 8:38,40). 28 See also John 8:32, “[. . . ] the truth will set you free”. 29 A relation between husband and wife could be closely described by Kant’s category of “reciprocity” (Wechselwirkung). Kant gives the “reciprocity” (“community”) of the parts of a body as an example (B 112). This could be compared with the following place from Genesis: “a man leaves his father and mother and clings to his wife, and the two become one body” (Genesis 2:24). We can also notice here that the community of “man and wife” is a sort of a new causal beginning, discontinuity with past. Let us mention that there are strong indications in John 4 that Jesus could be understood, in some true, spiritual sense, as a bridegroom (Jojko (2012)). However, according to Genesis, the community relationship between man and wife changed after the Fall to the rule of man over woman (“he [your husband] shall rule over you”, Genesis 3:16); this one-sided causality could be related to the Samaritan woman’s past husbands. The whole dialogue might be seen as a reversal of the Adam and Eve story of Genesis, that is, as a path leading back to the state of original unity with God. 30 In general, the Samaritan woman lives in the whole historical tradition of her people. For ex-
Concepts, Space-and-Time, Metaphysics |
73
This causality of a new start in John 4 has obviously its counterpart in Kant’s causality of freedom, but, in distinction, it is not reducible to an agent’s (self’s) “spontaneity” (“self-activity”, B 446) without dependence on God.³¹ The causality of John 4, again, leads to a contradiction with respect to possible ways of God’s presence, that is, between the immediate presence of God, independently of the contingencies of place and time, and the dependence of God’s presence on some (contingent) place and time (see the fourth antinomy above). Remark 2.1. In terms of the system of mutually irreducible notions of possibility, imagination, and conception (as described by J.-Y. Béziau (2016)), we could interpret the somewhat surprising transition from the Samaritan woman’s wish to get living water and eternal life to Jesus’ request to her to call her husband, in the following way: we assume that her wish is conceivable for her (she understands its meaning), and also imaginable (for instance, in some Pentateuchal picture³²), but there remains the question of possibility (aimed at by Jesus): she cannot receive eternal life while still being in causal dependence on the sensible temporal world (her past life). The solution is that she should make her wish independent of this temporal causal sequence by relating herself, in her self-knowledge (truth), to God. In application to Kant’s antinomy: we could conceive (by categories) what it means to act out of freedom and imagine it in some intuitive form, but the possibility of such a causality opens up, for Kant, only in the intelligible moral world (otherwise,
ample: “[. . . ] our father Jacob, who gave us this cistern and drank from it himself with his children and his flocks” (John 4:12); and later: “Our ancestors worshiped on this mountain [Gerizim] [. . . ]” (John 4:20). Cf. “The woman had found her security in the concept of her ancestors, but this is now transcended by the reference to ‘the Father’ [. . . ]. There is a progression from a very narrow and limited view of the exclusivity of one group to the liberating discovery that God is the Father of all [. . . ]” (Botha (1991), p. 152). Sometimes, the determination by one’s physical origin may be insurmountable: “Jesus himself testified that a prophet has no honor in his native place” (John 4:44). 31 For instance: “the idea of spontaneity, which could start to act from itself, without needing to be preceded by any other cause that in turn determines it to action according to the law of causal connection” (B 561); “a causality in our power of choice such that [. . . ] it might [. . . ] begin a series of occurrences entirely from itself ” (B 563). However, Kant points out: “The confirmation of the need of reason to appeal to a first beginning in the series of natural causes is clearly and visibly evident from the fact that (with the exception of the Epicurean school) all the philosophers of antiquity saw themselves as obliged to assume a prime mover for the explanation of motions in the world, i.e., a freely acting cause, which began this series of states first and of itself” (B 478). 32 E.g., Exodus 17:6: “Strike the rock, and the water will flow from it for the people to drink”. Cf. Num 20:11.
74 | Srećko Kovač the causality of freedom remains just ens rationis and ens imaginarium, i.e., a concept and an intuition without an object, see B 347–349). (d) The fourth antinomy (dependence in existence). The solution confirms Thesis 4, according to which the necessary being, God, is immediately present (exists) – not in some space and time of a sensible world, but “in Spirit and truth”: “the hour is coming, and is now here, when true worshipers will worship the Father in Spirit and truth” (John 4:23). In the sensible world, God is not immediately present, but probably only through a (long) intermediate sequence of the dependence on past events (coming, finally, to the beginnings of a religious tradition). Here we observe the similarity with Kant’s distinction between the “intelligible” necessity (“in Spirit and truth”), and “sensible” contingency. However, in John 4, not only is a necessarily existent being aimed at, but also its presence (in worship).³³ Thus the difference results between the presence of a necessary being in Spirit and truth, and a long “series of dependent existences” (B 587), eventually leading to some non-present (separate) necessary being. We see that Spirit should be “something” that is independent of the contingent sensible world, although, at the same time, it should be somehow active in that world. It is also clear that the truth about the sensible world is not as such part of the sensible world, although it is related to that world. The question about how God can be immediately present to us in Spirit and truth remains open and is addressed in the final section of the dialogue.
2.3 The highest reality We outline how the idea of God as the most real being (ens realissimum) (see (c) on page 62) can be traced back to the dialogue in John 4. In the Samaritan woman’s utterance on what she knows about the worship of God in Spirit and truth, the idea of God as the source of the knowledge of “everything” (of truth) is clearly present: I know that the Messiah is coming, the one called the Anointed; when he comes, he will tell us everything. (John 4:25).
33 For Kant, in distinction, “the necessary being would have to be thought of as entirely outside the series of the world of sense (as an ens extramundanum), and merely intelligible; this is the only way of preventing it from being subjected to the law of the contingency and dependence of all appearances” (B 589).
Concepts, Space-and-Time, Metaphysics |
75
The Messiah is “the one whom God has sent” and who “speaks the words of God” (John 3:34, cf. Dt 18:18,22). Thus the Messiah’s knowledge of “everything” stems from God, who, in some way, possesses truth about “everything”.³⁴ Since truth about everything is truth with respect to all possible predicates (according to the “principle of complete determination”, see (c) on page 62), God obviously possesses the totality of all concepts (predicates). Here, a Kantian reasoning can be applied according to which God is completely determined by the idea of the “possession of all reality” (Allbesitz der Realität), as the entity having all positive properties (the negative ones being defined by means of the corresponding positive ones), and thus should be conceived as ens realissimum (B 604). In distinction to Kant’s view, in John 4 God is not merely a regulative ideal (or a postulated being), not even a being that will be present only in the future (as in the Samaritan woman’s above-quoted statement, John 4:25), but is conceived as actually existing – present here and now, in the dialogue itself. This presence is indicated by the self-revealing words: “I am [eg¯o eimi]³⁵, the one who is speaking with you” (John 4:26).
3 Through dialogue to a metaphysical experience (a formalization) Summarizing the preceding section, it can be said that in John 4 a metatheoretical conceptual structure is present that is analogous to Kant’s system of transcendental ideas, one of the main differences being that these ideas should have actual denotation in religious experience instead of having a merely regulative or postulated role. Accordingly, the system of transcendental ideas should become a religiously based metaphysical system. This actual denotation of metaphysical
34 This “everything” is later reflected in the Samaritan woman’s words: “Come see a man who told me everything I have done. Could he possibly be the Messiah?” (John 4:28, cf. John 4:39). The Samaritan woman’s awareness that omniscience is grounded in God is confirmed by her wish to worship God after Jesus told her “everything she has done”; see the third antinomy above and John 4:18–20,29,39. 35 This is “an Old Testament self-designation of Yahweh” (New American Bible (2011), p. 1439, footnote ‡). “It [the affirmation of eg¯o eimi] was used to manifest the living presence of God who makes himself accessible to his people. Jesus using this title manifests the visible presence of God to the Samaritan woman and eventually to the Samaritans” (Varghese (2009), p. 134).
76 | Srećko Kovač concepts, as well as of concepts in general, should be grounded in their spiritual sense, as presented in John 4.³⁶ In the following, we describe in a formal way how a religious reality of metaphysical concepts is achieved through the dialogical interaction of agents (Jesus, the Samaritan woman). To that end, we use a modification of justification logic (stemming from Gödel (1938); see also Artemov (2001)), where the concepts that were observed above as being transformed from metatheoretical to metaphysical ones will be interiorized into an object theory.
3.1 System QJDR The vocabulary consists of individual constants c, d, e, c1 , . . . and individual variables x, y, z, x1 , . . .; predicate letters P ij ; =; ¬, →; the quantifier symbol ∀; term operation symbols +, ·, ! and genx . Individual terms are individual variables and constants, and complex terms (t + u), (t · u), !t, and genx (t), where t and u are individual terms. Formulas are of the shape Pt1 . . . t n , t = u, ¬ϕ, (ϕ → ψ), ∀uϕ, t : ϕ, with the meaning ‘t has/gives evidence that ϕ’, and t : : ϕ meaning ‘t wishes (requests) ϕ’. Symbols ∧, ∨, ↔ and ∃ are defined in the usual way. We will also informally use accommodated English words to facilitate the understanding of translations. ⊥ will stand for a contradiction, ϕ ∧ ¬ϕ. The axiomatic system QJDR is designed on the basis of justification logic systems QLP by Fitting (2008) and FOLP by Artemov and Yavorskaya (2011), without factivity (an analogue of modal axiom T), with the addition of = and wish operator : : , and with individual terms as evidence and wish terms. The axioms are: CPC a b Id Rg Sub JMon
classical propositional tautologies ∀xϕ → ϕ(t/x), t is substitutable for x in ϕ ∀x(ϕ → ψ) → (ϕ → ∀xψ), x ∈ / free(ϕ) x=x x = y → ∃z z : x = y ¬x = y → ∃z z : ¬x = y x = y → (ϕ(x) → ϕ(y)), ϕ is atomic x : ϕ → (x + y) : ϕ y : ϕ → (x + y) : ϕ
36 John 4 could be compared with John 3, where, in Jesus’ dialogue with Nicodemus, we cannot see Nicodemus succeeding to access the spiritual reality (Nicodemus remained dependent on his naturalistic and historicist knowledge) (see Kovač (2015a)).
Concepts, Space-and-Time, Metaphysics |
JK J4 J∀ DMon, DK, D∀ DJK JDK DJ4
77
x : (ϕ → ψ) → (y : ϕ → (x · y) : ψ) x : ϕ → !x : x : ϕ t : ϕ → genx (t) : ∀xϕ, x ∈/ free(t) like JMon, JK, J∀, respectively, with : : for : x : : (ϕ → ψ) → (y : ϕ → (x · y) : : ψ) x : (ϕ → ψ) → (y : : ϕ → (x · y) : : ψ) x : : ϕ → !x : x : : ϕ
as well as the following special axioms: SA1J SA1D SA2 SA3 SA4 SA5 SA6
InTruth(x) → (x : ϕ → ϕ) InTruth(x) → ¬x : : ⊥ InTruth(x) → InTruth(!x) InTruth(t) → InTruth(genx (t)), with x ∈/ free(t) (InTruth(x) ∧ InTruth(y)) → InTruth(x · y) InTruth(x + y) → (InTruth(x) ∨ InTruth(y)) special axioms including meaning postulates about non-logical symbols (to be introduced, sometimes implicitly, during the formalization of the dialogue of John 4).
Rules are modus ponens (MP), universal generalization (UG), and axiom justification (AJ): if ⊢ ϕ, then ⊢ c : ϕ, where ϕ is an axiom, and c a justification constant – according to some constant specification function CS, which assigns a justification constant to each axiom (in our translation of John 4, the assigned constants will be, informally, j and s). In addition, some facts (possibly evidence and wishes) will appear as proof lines in the translation of the reasoning in the dialogue of John 4. In some places with complex agents involving only one basic agent t, we will use an indexed expression (t)n for short. Remark 3.1. A unified definition of a model and a variable assignment can be proposed, with some basic features of the definition of a Mkrtychev model in FOLP (Fitting (2014); for propositional logic, see Mkrtychev (1997)), but extended with the identity relation and functions for wishes and requests, as well as including evidence and wish agents together with their complexes into the domain of a model (see Fitting’s semantics of QLP for evidence in Fitting (2008), and the causal semantics for QCGO in Kovač (2015b)). Model, M, and variable assignment, v, make n-tuple ⟨D* , I, v, Ev, Wish⟩, where (a) D* is a set built on a basic set of individu-
78 | Srećko Kovač als (D) and closed under evidence and wish operations,³⁷ (b) I is an interpretation function mapping individual constants to individual objects of D, and mapping relation symbols (predicate letters, term operation symbols) to relations on D* , (c) v is a variable assignment mapping each variable to a member of D* , so that now the denotation of a term, JtK (which is short for JtKM v ), is I(c) or a complex object – !J u K, Ju1 K+Ju2 K, Ju1 K·Ju2 K, genx (JuK) – or v(x), depending on whether t is a constant, a complex term, or a variable; finally, (d) evidence and wish functions Ev and Wish map each member of D* to a subset of formulas, in analogy with the axioms: 1. 2.
3. 4.
ϕ ∈ Ev(JtK) or ϕ ∈ Ev(JuK) =⇒ ϕ ∈ Ev(JtK+JuK), ϕ ∈ Wish(JtK) or ϕ ∈ Ev(JuK) =⇒ ϕ ∈ Wish(JtK+JuK), ϕ → ψ ∈ Ev(JtK) & ϕ ∈ Ev(JuK) =⇒ ψ ∈ Ev(JtK·JuK), ϕ → ψ ∈ Wish(JtK) & ϕ ∈ Wish(JuK) =⇒ ψ ∈ Wish(JtK·JuK), ϕ → ψ ∈ Ev(JtK) & ϕ ∈ Wish(JuK) =⇒ ψ ∈ Wish(JtK·JuK), ϕ → ψ ∈ Wish(JtK) & ϕ ∈ Ev(JuK) =⇒ ψ ∈ Wish(JtK·JuK), ϕ ∈ Ev(JtK) =⇒ t : ϕ ∈ Ev(!JtK) (positive introspection), ϕ ∈ Wish(JtK) =⇒ t : : ϕ ∈ Ev(!JtK) (positive introspection), ϕ ∈ Ev(JtK) =⇒ ∀xϕ ∈ Ev(genx (JtK)), x ∈/ free(t) ϕ ∈ Wish(JtK) =⇒ ∀xϕ ∈ Wish(genx (JtK)), x ∈/ free(t).
The satisfaction of a formula is defined classically for atomic, compound and quantified formulas. For evidence and wish formulas, the conditions are as follows: (a) (b)
M |=v t : ϕ ⇔ ϕ ∈ Ev(JtK), M |=v t : : ϕ ⇔ ϕ ∈ Wish(JtK).
In principle, it should not be difficult to define the denotation of InTruth and of the non-logical relation symbols used below.
3.2 Translation of the dialogue We will now translate some characteristic moments of the dialogue. Instead of formal descriptive symbols, we will mainly use abbreviations which by themselves indicate their meaning; j will denote Jesus, s the Samaritan woman, w physical (transient) water, and w′ true (eternal) water. In the translation, we will focus on the interconnection of agents that is being established during the process of resolving contradictions and of increasing knowledge. Numbers in square brackets
37 In semantic metalanguage, the evidence and wish operations will be denoted by bolded evidence and wish operation symbols.
Concepts, Space-and-Time, Metaphysics |
79
will indicate the reference of a reply or an opposition to a previous proposition in the dialogue.
Part 1 At the beginning of the conversation, Jesus actually asked for true water (w′ ), but this was not properly understood by the Samaritan woman. We will, first, show Jesus’ request and beliefs (in a slightly simplified way) in the sense that the Samaritan woman thought they were meant, and thereafter we will express that they are being perceived so by the Samaritan woman:³⁸ 1 2 3 4 5 6
j : : Gives(s, j, w) j : (j : : Gives(s, j, w) → (¬j : : Gives(s, j, w) → ⊥)) (j)g : ((Jew(j) ∧ Samaritan(s)) → ¬j : : Gives(s, j, w)) j : (Jew(j) ∧ Samaritan(s)) ((j)g · j) : ¬j : : Gives(s, j, w) ((j · j) · ((j)g · j)) : : ⊥
fact AJ SA6,AJ fact 3, 4 JK, [1] 1, 2, 5 JK, JDK
As mentioned above, this is not quite what Jesus meant – it is how the Samaritan woman understood his words (the numerals in parentheses indicate the respective sentences of the formalization above): 7 8 9 10 11 12
s : ((1) ∧ (2) ∧ (3) ∧ (4)) (s)h : (((1) ∧ (2) ∧ (3) ∧ (4)) → (6)) ((s)h · s) : (6) (s)i : ((6) → ¬InTruth(((j · j) · ((j)g · j))) ((s)i · ((s)h · s)) : ¬InTruth(((j · j) · ((j)g · j))) ((s)j · ((s)i · ((s)h · s))) : ¬InTruth(j)
fact 1-6 AJ 7, 8 JK SA1D, AJ 9, 10 JK 9 SA4, AJ
We take that Jesus is aware of 12 (that the Samaritan woman, in her reasoning, does not believe him at the time). He therefore introduces another viewpoint by advancing his belief that what should happen is Gives(j, s, w′ ), referring to “true
38 For simplicity, the translation is in some places reductive on the ground of context. For example, ‘to give a drink’ is translated as ‘to give water’, and the Samaritan woman’s question on how Jesus can ask her for a drink is translated simply as the assumption that Jesus is actually not asking her (because he should not ask her) to give him water. See a detailed interpretation, for example, in Botha (1991), pp. 115–122.
80 | Srećko Kovač water”. However, this is understood by the Samaritan woman again as Gives(j, s, w), i.e., as referring to sensible water. Let us now assume that j : Gives(j, s, w′ ) (1) as well as s : ¬Gives(s, j, w) (2) hold. Since we take that j has evidence about all logical axioms, and since we thus get j : (Gives(j, s, w′ ) → (Gives(j, s, w′ ) ∨ ¬Gives(s, j, w))), so, starting from (1), (j · j) : (Gives(j, s, w′ ) ∨ ¬Gives(s, j, w)) follows. In an analogous way, starting from (2), we can derive (s · s) : (Gives(j, s, w′ ) ∨ ¬Gives(s, j, w)). From both sentences we obtain ((j · j) + (s · s)) : (Gives(j, s, w′ ) ∨ ¬Gives(s, j, w)) by JMon. Hence, we derive: (j : Gives(j, s, w′ ) ∨ s : ¬Gives(s, j, w)) → ((j · j) + (s · s)) : (Gives(j, s, w′ ) ∨ ¬Gives(s, j, w)). In addition, an application of the proof example by Artemov (2001), p. 10, to our context shows that evidence for j : Gives(j, s, w′ ) ∨ s : Gives(j, s, w′ ) includes the positive introspection of s and j, that is (j : Gives(j, s, w′ ) ∨ s : Gives(j, s, w′ )) → ((j · !j) + (s · !s)) : (j : Gives(j, s, w′ ) ∨ s : Gives(j, s, w′ )).
Part 2 By a reasoning similar to that in Part 1, we obtain the following sequence: (1) j : : Gives(j, s, w) (fact), (2) (j)k : (Gives(j, s, w) → HasBucket(j)) (SA6, AJ), (3) ((j)k · j) : : HasBucket(j), (4) j : ¬HasBucket(j) (fact), (5) j : (HasBucket(j) ′ ′ → (¬HasBucket(j) → ⊥)), (6) ((j · ((j)k · j)) · j) : : ⊥. Like in Part 1, ((s)i · ((s)h · s)) : ¬InTruth(((j · ((j)k · j)) · j)) holds, and thus, for some complex built of s, an evidence for ¬InTruth(j) again results (cf. SA4). Therefore, Jesus introduces a more explicit distinction between w and w′ , which is noticed (although still not fully understood) by the Samaritan woman: 1 2 3 4 5 6
j : : Gives(j, s, w′ ) j : (Drinks(s, w′ ) → NeverThirsty(s)), j : (Neverthirsty(s) → Drinks(s, w′ )) s : : Neverthirsty(s) (j · s) : : Drinks(s, w′ ) s : (Drinks(s, w′ ) → Gives(j, s, w′ )) (s · (j · s)) : : Gives(j, s, w′ )
fact fact fact 2, 3 JDK fact 4, 5 JDK, [1]
Concepts, Space-and-Time, Metaphysics |
81
However, the realizability of (s ·(j · s)) : : Gives(j, s, w′ ), that is, of getting “a spring of water welling up to eternal life”, depends on a further ascent, to be initiated by Jesus, in the Samaritan woman’s knowledge.
Part 3 1 2 3
s : j : : ∃x(Call(s, x) ∧ HusbandOf (x, s)) s : ¬∃xHusbandOf (x, s) j : HasDoneS(s)
4
∀y(HasDoneS(y) → ¬∃xHusbandOf (x, y)) (j)l : (¬∃xHusbandOf (x, s) ∧ s : ¬∃xHusbandOf (x, s))
5 6 7 8 9 10 11 12
s : j : HasDoneS(s) j : HasDoneS(s) → Prophet(j) s : (j : HasDoneS(s) → Prophet(j)) (s · s) : Prophet(j) ∀x(Prophet(x) → InTruth(x)) (s · s) : (Prophet(j) → InTruth(j)) ((s · s) · (s · s)) : InTruth(j)
fact fact, [1] fact, “he told me everything I have done”, John 4:39 the meaning of HasDoneS, SA6 3, 4, assuming j : (2) as a fact, [2] 3 fact SA6 7 AJ 6, 8 JK SA6 10 AJ* 9, 11 JK, [12] of Part 1, cf. Part 2
(* In line 11, we apply AJ (for s) and ∀a to line 10, and hence, again by means of AJ, we get s : (∀x(Prophet(x) → InTruth(x)) → (Prophet(j) → InTruth(j)) as an intermediate step). Of course, the beliefs of lines 9 and 12 are true: 13 14
Prophet(j) InTruth(j)
3, 7 10, 13
For 13, see Dt 18:18 and cf. with John 1.
Part 4 Thus, the Samaritan woman concludes that, in some place, God should be worshiped:
82 | Srećko Kovač 1 2 3 4
(s)l : (∃xProphet(x) → ∃xWorshipedIn(god, x)) (s · s) : (Prophet(j) → ∃xProphet(x)) ((s · s) · (s · s)) : ∃xProphet(x) ((s)l · ((s · s) · (s · s))) : ∃xWorshipedIn(god, x)
SA6, AJ ∀a, tautology, AJ
9 of Part 3 JK 1, 3 JK
However, there is controversy about the place of worship: 5 6 7 8 9
samaritans : WorshipedIn(god, gerizim) jews : WorshipedIn(god, jerusalem) s : (WorshipedIn(god, gerizim) → (WorshipedIn(god, jerusalem) → ⊥)) ((s · samaritans) · jews) : ⊥ ¬InTruth(((s · samaritans) · jews))
fact, [4] fact, [4] fact 5–7 JK 8 SA1J
According to the Samaritan woman it cannot be that both Samaritans and Jews are right regarding the place of worship. Thus Jesus points to worship in Spirit and truth as the solution to the controversy (see the next, final, part).
Final Part 1 2 3 4
j : ∀x(WorshipedIn(god, x) ↔ (InTruth(x) ∧ InSpirit(x))) s : ∃xTheAnointed(x) s : ∀x(TheAnointed(x) → (InTruth(x) ∧ InSpirit(x))) j : ∀x(TheAnointed(x) → x = j)
fact fact (∃x referring also to the future), [1] fact, [1] (cf.: x “will tell us everything”, John 4:25) fact, [2]
We derive some consequences not explicitly stated in the Gospel text: 5 6 7 8 9
(s)n : (∀x(TheAnointed(x) → x = j) → (∃xTheAnointed(x) → TheAnointed(j))) ((s)n · j) : (∃xTheAnointed(x) → TheAnointed(j)) (((s)n · j) · s) : TheAnointed(j) ((s · s) · (((s)n · j) · s)) : (InTruth(j) ∧ InSpirit(j)) ((j)m · ((s · s) · (((s)n · j) · s))) : WorshipedIn(god, j)
first-order logic, iterated JK 4, 5 JK 2, 6 JK 3, 7 JK* 1, 8 JK* , [1]
(* See analogous note * for line 11 of Part 3.) Line 9 includes some propositional logic as evident to j. Of course, according to J4 the following is derivable: !(((s)n · j) · s) : (((s)n · j) · s) : TheAnointed(j) as well as !((j)m · ((s · s) · (((s)n · j) · s)) : ((j)m ·
Concepts, Space-and-Time, Metaphysics |
83
((s · s) · (((s)n · j) · s)) : WorshipedIn(god, j). In addition, everything that was in the proof until line 9 stated for s holds for j, too, from where and from line 14 of Part 3, it follows (by SA1J): 10 11 12
∀x(TheAnointed(x) → x = j)
InSpirit(j) WorshipedIn(god, j)
see 4 see 8 see 9
Thus, non-formally expressed, the result is that Jesus (j) is a true, spiritual self (“I am”) in which God should be worshiped. As the Samaritan woman returned to her town, she told the people about what had happened; thus she may have reasoned in more general terms. For example, her reasoning from Part 2 might also have been as follows: from her generalized wish s : : (HumanInTown(x) → Neverthirsty(x)), and from j : (Neverthirsty(x) → Drinks(x, w′ )) (cf. Part 2, line 2) it follows (with a bit of s’s evidence of propositional logic) that ((s · s) · j) : : (HumanInTown(x) → Drinks(x, w′ )). Obviously, from
s : (Drinks(x, w′ ) → Gives(j, x, w′ ))
we can derive (again, on the ground of s’s evidence of propositional reasoning) ((s · ((s · s) · j)) · s) : : (HumanInTown(x) → Gives(j, x, w′ )). Now, according to D∀, we conclude: genx (((s · ((s · s) · j)) · s)) : : ∀x(HumanInTown(x) → Gives(j, x, w′ )). In distinction to John 3 (a conversation with Nicodemus), we see how through the complexity of the agents’ interconnections and during the progress of resolving contradictions and of gaining knowledge, the agents become “reborn” in the “water” of the dialogue and “in Spirit”. At the same time, we see how metaphysical concepts, especially the concepts of self, world and God, if approached from a religious viewpoint, could obtain a sort of objective reality beyond the realm of sensible objects.
84 | Srećko Kovač Acknowledgement I am grateful to Mirosław Szatkowski for the invitation to the conference God, Time, Infinity, Warsaw, September 22–24, 2015, where the first version of the paper was presented. This work has been supported by the Croatian Science Foundation under project IP-2014-09-9378.
Bibliography Achourioti, T. and van Lambalgen, M. (2011), “A Formalization of Kant’s Transcendental Logic”, in Review of Symbolic Logic: 4, 254–289. Artemov, S. (2001), “Explicit Provability and Constructive Semantics”, in Bulletin of Symbolic Logic: 7, 1–36. Artemov, S. and Yavorskaya (Sidon), T. (2011), “First-order Logic of Proofs”, in CUNY Ph.D. Program in Computer Science, TR-2011005. Béziau, J.-Y. (2016), “Possibility, Imagination and Conception”, in Principios: 23, 59–95. Botha, J. E. (1991), Jesus and the Samaritan Woman: A Speech-Act Reading of John 4:1-42, Leiden: E.J.Brill. Fitting, M. (2008), “A Quantified Logic of Evidence”, in Annals of Pure and Applied Logic: 152, 67–83. Fitting, M. (2014), “Possible World Semantics for First-Order LP”, in Annals of Pure and Applied Logic: 165, 225–240. Gödel, K. (1947/1964), “Cantor’s Continuum Problem”, in Kurt Gödel, Collected Works, Vol. 2, edited by S. Feferman et al., Oxford: Oxford University Press, 1990, 254–270. Gödel, K. (1938), “Vortrag bei Zilsel”, in Kurt Gödel, Collected Works, Vol. 3, edited by S. Feferman et al., Oxford: Oxford University Press, 1995, 86–113. Jojko, B. (2012), Worshiping the Father in Spirit and Truth, Rome: Gregorian and Biblical Press. Kant, I. (1910–), Kant’s Gesammelte Schriften, edited by the Königlich Preussische Akademie der Wissenschaften, Berlin: Reimer; and Berlin and Leipzig: de Gruyter. Also Kants Werke, I–IX, Berlin: de Gruyter, 1968. Kant, I. (1781/1787) Kritik der reinen Vernunft, in Kant (1910–), Vol. III–IV, 1st ed. 1781 (A), and 2nd ed. 1787 (B). Kant, I. (1783) Prolegomena zu einer jeden künftigen Metaphysik, in Kant (1910–), Vol. IV. Kant, I. (1998) Critique of Pure Reason, translated and edited by P. Guyer and A. W. Wood, Cambridge [UK]: Cambridge University Press. Kant, I. (1997) Prolegomena to Any Future Metaphysics, translated by P. Carus, Chicago, La Salle: Open Court. Kovač, S. (2011), “Logika i vjera”, in Suvremena znanost i vjera : Contemporary Science and Faith, edited by Z. Primorac, Mostar, Ljubljana, etc.: Fakultet prirodoslovno-matematičkih i odgojnih znanosti Sveučilišta u Mostaru, 69–84. Kovač, S. (2015a) “Logic and Truth in Religious Belief”, in God, Truth, and Other Enigmas, edited by M. Szatkowski, Berlin, etc.: de Gruyter, 119–132.
Concepts, Space-and-Time, Metaphysics |
85
Kovač, S. (2015b) “Causal Interpretation of Gödel’s Ontological Proof”, in Gödel’s Ontological Argument: History, Modifications, and Controversies, edited by K. Świętorzecka, Warsaw: Semper, 163–201. Mkrtychev, A. (1997), “Models for the Logic of Proofs”, in Logical Foundations of Computer Science, edited by S. Adlan and A. Nerode, Berlin, Heidelberg, etc.: Springer, 266–275. Nestle-Aland Novum Testamentum Graece (2012), Stuttgart: Deutsche Bibelgesellschaft. New American Bible, Revised Edition (2011), Charlotte, NC: Saint Benedict Press. Also: New American Bible, 2002, http://www.vatican.va/archive/ ENG0839/_INDEX.HTM Novakovic, L. (2013), “Jews and Samaritans”, in The World of the New Testament, edited by J. B. Green and L. M. McDonald, Grand Rapids: Baker Academic, 207–216. Tiles, M. (2004), “Kant’s Logic”, in Handbook of the History of Logic, Vol. 3, edited by D. Gabbay and J. Woods, Amsterdam etc.: Elsevier, 85–130. Varghese, J. (2009), The Imagery of Love in the Gospel of John, Rome: Gregorian and Biblical Press.
Zbigniew Król
Basic Intuitions Concerning the Concept of Infinity in Mathematics from a Historical and Theological Point of View Thus the conquest of actual infinity may be considered an expansion of our scientific horizon no less revolutionary than the Copernican system or than the theory of relativity, or even of quantum and nuclear physics. Fraenkel and Levy (1976), p. 40.
1 Introductory remarks The history of human understanding of itself and reality consists in many anonymous, however, fundamental discoveries. There had to have been someone who noticed that there were such “things” as time, space or infinity. It is impossible to experience them through the senses – sight, hearing, smell or touch. These remarkable insights are works of unknown geniuses who still guide and determine our everyday, as well as scientific, life. The concept and problem of infinity is still far from having been fully analyzed. However, we know much more about infinity now. The intuition of infinity consists of many factors which determine the possible directions of further analysis. We have the intuitions of infinitely large and small quantities,¹ actual and potential infinity, intuitions of infinite series and infinitely divisible quantities, as well as intuitions of many actually infinite objects: infinite straight line, space, surface, the set of natural numbers etc. We can imagine infinity in space and time, e.g. the idea of eternity, or in perfection. Such ideas, in a natural way, are directed toward the concept of God, who is often considered as the source and cause of every kind of “human” infinity. One can say much more: the analyses of the consistency of the concept of God, who is considered as omnipotent, infinitely powerful, free, eternal etc., are the main sources for the development of the concept of infinity, not only from the historical point of view.
1 Cf. for instance non-standard analysis; see Robinson (1966). https://doi.org/9783110594164-006
88 | Zbigniew Król
2 The development of the concept of infinity The ways of understanding the concept of infinity have been changing through the ages. The concept was, at first, interpreted with the help of the concept of a limit (peras) and it was defined as apeiron, i.e. non-limited, unlimited, undetermined etc. Something was infinite when it lacked an ascertainable limit. Thus, infinite series are infinite because there is no limit in increasing them infinitely. It seems that the first kind of analyzed infinity was potential infinity. In the same way, lines are infinite because one can divide them without any limits, i.e. any part of the divided continuous quantity can be divided later into smaller and smaller parts, or one can produce them without any limits. This type of infinity belongs to the type of undefined, undetermined quantities. Therefore, Plato, in his unwritten doctrine, Philebus or Parmenides, considered that everything is composed of two factors: the two Highest Principles, i.e. the One and the (Undetermined), the Dyad of that which is Great and Small. The One imposes some limits on the Dyad. The highest principles are also present in ancient mathematics. For instance, every mathematical entity in Euclid’s Elements is composed of some limiting factors and of something undetermined; see, the lines or figures in Elements. The lines without endpoints were called “infinite”. However, the lines or figures are finite but of undefined size. They play the role of geometrical variables; see also the opinion of Euler in his (Euler, 1797).² As I argue in my book, Platonism and the Development of Mathematics. Infinity and Geometry (Król (2015)) actually infinite objects were actually known in Antiquity. However, they were excluded from mathematics and geometry. For instance, there is no infinite line, surface or space in Euclid’s Elements or by Archimedes. The problem of operating with the wholes (concepts) of actually infinite scopes is the main reason why the ancient authors did not consider the problem of a possible change of the basic line in Book X of Elements. The only two exceptions to ancient finitism were: 1.
Proclus’ proof that to a given line there is only one (infinite) parallel line different from the given line;³
2 Cf. also Euler (1990). 3 Only this first example demonstrates the unique antique use of actual infinity in a mathematical proof. Proclus, explaining theorem I.30, i.e., that straight lines parallel to the same straight line are also parallel to one another, says (contrary to Euclid): “For we must think of parallel lines as produced indefinitely, and AH when produced coincides with HB; it is therefore the same as it, and not another line. Therefore all the parts of a parallel line are themselves parallel to the straight line to which it is parallel, both to the whole of it and to its parts. Thus AH is parallel to
Basic Intuitions Concerning the Concept of Infinity in Mathematics |
2.
89
an attitude taken by Apollonius in his Conics.⁴ (This last case is not as evident as the first one.)
It is possible to indicate basic historical changes in the understanding of the concept of infinity. At the very beginning, infinity is considered as potential infinity and it is analyzed with the use of the concept of “unlimitedness”. Some other concepts or intuitive convictions are also essential. For instance, that the whole is greater than a part of it or that the sum of infinitely many quantities which are “greater than nothing” is infinitely great. There are also some beliefs which are non-verbal, i.e., that there is only one infinity in every kind.
3 God’s point of view and the emergence of actual infinity in mathematics The so-called, God’s point of view was very important in the process of the emergence of actual infinity in mathematics. Philosophers started their analysis from God’s point of view, i.e. they considered logical situations in which the assumptions concerned the omnipotent, eternal and all-knowing subject of cognition. For a limited human being, it was impossible to actually divide a continuum to infinity during a finite lifetime etc. However, it was possible for him to analyze the
KD and HB to CK; for when produced indefinitely they remain nonsecant.” (Cf. Proclus (1987), p. 294). He also proves the uniqueness of the parallel line through the given point and parallel to the given line: “From the same point two perpendiculars cannot be drawn to the same straight line, nor through the same point can two parallels be drawn to the same straight lines.” (Proclus (1987), p. 296). However, Proclus’ statement is in contradiction with Euclid’s explicit proofs from Elements. Proclus mixes the meanings of the actually infinite line and the undefined or undetermined line: there is really only one parallel and undefined line; i.e., a line without limited points, and many finite parallel lines. He is also not firm in the above statement, and contradicts himself in many places; cf. p. 303 (“For a parallelogram is formed by the equal and parallel lines”) or p. 312. 4 In contradiction to Euclid, Archimedes and others who, in their definitions and propositions, use only determined finite cones, Apollonius defines an indefinite conic line and conic surface, and discerns these objects from every given cone: “If a straight line indefinite in length, and passing through a fixed point, be made to move round the circumference of a circle which is not in the same plane with the point, so as to pass successively through every point of that circumference, the moving straight line will trace out the surface of a double cone, or two similar cones lying in opposite directions and meeting in the fixed point, which is the apex of each cone.” (Heath (1896), p. 1). Based on Apollonius’ Conics, it is necessary to differentiate between the actually infinite line or surface and undefined in length, i.e. an unlimited line and surface respectively.
90 | Zbigniew Król logical consequences of such premises as if a man could exist like God. This last point is exemplified by the medieval discussions concerning the concept of God’s omnipotence or some properties of the continuum. Without the consideration of God’s properties and divine possibilities or without the application of God’s cognitive perspective, the use of actually infinite objects in mathematics would not have been possible. One of the most well-known and broadly discussed argument in the Middle Ages was Henry of Harclay’s,⁵ who lived a little later than Nicole Oresme. The case is reported by William of Alnwick in his Determinationes, next quoted by Adam Wodeham in Tractatus de indivisibilibus:⁶ God actually sees or knows the first beginning point of a line, and any other point which it is possible to pick out in the same line. Therefore, either [i] God sees that, in between this beginning point of the line and any other point in the same line, a line can intervene, or [ii] not. If not [i.e., (ii)], then he sees point immediate to point, which is what we propose. If so [i.e., (i)], then, since it is possible to assign points in the intermediate line, those points will not be seen by God, which is false. This consequence is clear, for according to what we have posited a line falls between the first point and any other point (of the same line) seen by God, and consequently, there is some midpoint between this point and any other point seen by God. Therefore this midpoint is not seen by God.
Three main groups of theories concerning a continuum were formulated in the 14th century as a consequence of such discussions: 1. there are no points at all being the parts of a continuum, 2. a continuum is composed of infinitely many parts, each of which is also a continuum; an example are the views of Gregory of Rimini.⁷ A continuum is composed of infinitely many indivisible parts, for example, points. More information about this topic and about the emergence of actual infinity in mathematics can be found in my book. The works of Nicole Oresme (c. 1320–1387) are the most important from the mathematical point of view in this process.⁸ We find in his works a very mature and conscious use of the concept of actual infinity, both in geometry and in arithmetic. Nicole Oresme applies the infinite concepts in these two domains. The first use is exemplified by the summation of some infinite series. The second attitude can be found in his theorems concerning the commensurability and incommensurability of circular motions. However, the geometrical way of thinking is also present in arithmetic because he considers arithmetical theorems mainly from the geomet-
5 6 7 8
The reader can find more examples in Grant (1982). See, Cross (1998), p. 89. Analysis and literature on the subject, see, Cross (1998), p. 3. In this point, I adopt the results of my considerations from Król (2015), Chapter 13.
Basic Intuitions Concerning the Concept of Infinity in Mathematics |
91
rical point of view. This attitude is exceptional because, at almost the same time, purely logical and arithmetical reasoning are present, for instance, in the works of the Oxford’s calculatores. In Oresme’s works, one can find for the first time the prototypes of infinite lines, surfaces and also infinite, three-dimensional space; see, Król (2015), Chapter 13. The word “summation” of some infinite series of numbers can be misleading because, in reality, Oresme grounds the summation on a division of the given finite magnitudes into an actually infinite number of proportional parts. Oresme’s main mathematical innovation is the adoption of God’s point of view: he performs the aforementioned divisions and proofs, assuming that such a division is made actually in infinity. This small step is absolutely revolutionary from the mathematical point of view. The last six questions of his work Tractatus de configurationibus qualitatum et motuum, III.viii–III.xiii, contain constructions and proofs with the use of actually infinite geometrical objects and infinite series. Oresme applies the actually infinite series only for geometrical problems and they correspond to some easily imaginable geometrical objects. For instance, in question III.viii, he considers the following problem: “A finite surface can be made as long as we wish, or as high, by varying the extension without increasing the size.”⁹ In order to demonstrate how it is possible also in the case of a surface, he considers two one-square foot identical squares. He divides both of them into actually infinitely many proportional parts (sub-surfaces have areas equal to 1/2, 1/4, 1/8, . . . , which are designated by Oresme as E, F, G, etc.), the sum of which is – in an easily visible way – equal to the one foot square in both cases. Then, he uses the actually infinite number of these parts of the second square and stands them on top of the proportional parts of the first square: first he places 1/2 of the second square on top of the first one, then, 1/4 of the remaining half of the second square he places on top of the previous 1/2, etc., and all the parts have one collinear side which extends to infinity. He also demonstrates that the area of such a “stepped” actually to the infinity figure which actually contains infinitely many parts of determined, finite areas, is also finite, i.e. it is equal to a two square foot square. From the modern point of view, Oresme geometrically sums up the actually infinite series of the form: 1 + 1/2 + 1/4 + . . . + 1/(2n) + . . . = 4(1/2) = 2. He comments: Then upon this whole let the second part, namely F, be placed, and again upon the whole let the third part, namely G, be placed, and so on for the others to infinity. When this has
9 Cf. Clagett (1968) , p. 413.
92 | Zbigniew Król been done (“Quo facto”), let the base line AB be imagined as being divided into parts continually proportional according to the ratio of 2 to 1. . . . (Clagett (1968) p. 415)
This very simple step is revolutionary, “when this has been done”, enables the strict analysis of actually infinite geometrical objects. The summations, in every case (cf. the next questions of Part III De configurationibus. . . ) are based on some divisions into actually infinitely many proportional parts of some finite magnitudes. As I demonstrate in my book, the discovery and use of actually infinite objects in mathematics is a very complicated historical process which reaches its akmé in the works of Nicole Oresme, Cavalieri, Torricelli, Newton and Euler. However, Georg Cantor said much more about the nature of infinite collections than anybody before him. Infinite objects in mathematics, especially in geometry, were in common use in his times as a result of the application of God’s point of view. However, infinity, even in Cantor’s times, was considered as God’s own essential property and as a quality which can help to discern created things from eternal God. Actual infinity has been reserved for God itself in Christian theology since the Middle Ages and even since ancient times.¹⁰ It is a well-known fact that Cantor believed that the theory of transfinite numbers was revealed to him by God. Cantor was aware of the philosophical implications of this theory. He speaks about it in the Introduction to Grundlagen einer allgemeinen Mannigfaltigkeitslehre. Opposition to Cantor’s Paradise among mathematicians (Kronecker, Brouwer, Poincaré) interplays with the voices of opponents arguing from the theological point of view (Neo-Thomists, cardinal Johannes Franzelin).¹¹ The presence of actually infinite collections and other infinite objects can be reconciled with traditional Christian theology only if the objects of mathematics are some fictions or are “imaginary” in the sense of old medieval theories. In the same way and because of theological obstacles, Descartes resigned from the concept of actual infinity in mathematics; cf. Król (2015). Descartes maintains the ancient difference between “infinite” and “indeterminate” things. “Infinite”, in all aspects, is only God because He is “perfect”, i.e. there is no possibility to be completed (in an aspect) for Him. The other things can be completed and, therefore, they are “indeterminate” and not perfect. Descartes
10 Cf. Davenport (1997), pp. 263–295. 11 The reader can find more information together with the relevant sources in Dauben (1977, 1979, 2004).
Basic Intuitions Concerning the Concept of Infinity in Mathematics |
93
explains the difference between the terms infinitus and indeterminatus in his Principles of Philosophy, Chapters XXVI. However, he adopts God’s point of view in many places and considers actually infinite objects as well as collections of points; cf. Król (2015), Chapter 14.2. It means that his methods are Platonic.
4 Infinity in modern set theory From this perspective, it is interesting to analyze what God’s point of view in modern set theory relies on, i.e., what are Platonic methods in reference to the concept of infinity. In the description of the methods, I omit a rather well-known point concerning the non-predicativity (“circularity”) of the axiom of infinity. In modern mathematics, the problem of infinity in set theory seems to be one of the most important. There are two basic strategies of the construction of an infinite set: upward and downward. The first is usually applied in a universe of well-founded sets and creates an infinite “tower” of sets under the roof of an infinite limit-set. The downward strategy enables the creation of non-well-founded infinite sets, e.g., a Mirimanoff-like infinite series of nested sets located inside a “bottomless pit”. According to the first strategy, which is usually formalized within the frames of ZF(C) or NBG etc., the axiom of infinity can have many formulations. However, the basic ones are as follows: ∃z[(∃x ∈ z).∀y.¬(y ∈ x) ∧ (∀x ∈ z)(∃y ∈ z).x ∈ y]; ∃z[(0 ∈ z) ∧ ∀x.(x ∈ z → {x} ∈ z)];¹² ∃z[(0 ∈ z) ∧ ∀x.(x ∈ z) → (x ∪ {x}) ∈ z];¹³ ∃z[(0 ∈ z) ∧ (∀xy.(x ∈ z ∧ y ∈ z → x ∪ {y} ∈ z)]; ∃z∃x.(x ⊂ z ∧ x ̸ = z ∧ x ÷ z); (there is a reflexive set, “x ÷ z” — “there is a one-to-one correspondence between sets x and z”;¹⁴ A.6. There is an infinite set (in Bernays’ sense);¹⁵ A.7. ∃z ̸ = 0[∀x.(x ∈ z → ∃y.y ∈ z ∧ x ⊂ y ∧ x ̸ = y).¹⁶ A.1. A.2. A.3. A.4. A.5.
12 Cf. Zermelo (1908), pp. 261–281. 13 Cf. Neumann (1923), pp. 199–208. 14 This is, obviously, Dedekind’s definition. 15 Cf. Bernays (1958), p. 150. Bernays uses the formal statement which asserts that the class of natural numbers is a set, i.e., “is representable” in his wording. From this and Definition 1 follows the existence of an infinite set. 16 Cf. Bernays (1958), p. 150. I have slightly modified the original version of the axiom which was originally introduced by J. von Neumann and K. Gödel.
94 | Zbigniew Król On the other hand, one can develop many concepts, such as the concepts of equinumerosity, ordinal number, transfinite induction, natural number, numeral or finite set, all of these without any concept of an infinite set.¹⁷ Therefore, the next possibility to grasp the concept of infinity in set theory is to define an infinite set as a set which is not finite. Def.1. Inf(z) ≡ ¬Fin(z)
In the system of Bernays, the existence of an infinite set follows from axiom A.6. It appears that the Zermelo’s axiom A.2 follows from A.7. In fact, these axioms are equivalent; cf. Bernays (1958), pp. 148–150. Moreover, all the above axioms are equivalent to one another ¹⁸ on the basis of some other axioms of ZF, i.e. the axioms of Pairing, Union, Power-Set, the axiom schema of Subsets and Replacement but without the axioms of Foundation and Choice.¹⁹ We also know that the axioms of infinity are independent from the other axioms of set theory ZF(C). The above axioms and definitions of an infinite set are non-predicative, i.e., they are circular.²⁰ From the technical point of view, it means that we cannot construct an actually infinite set from the previously given finite sets or prove its existence from other axioms. One can simply postulate and accept the existence of an infinite set. Obviously, many other versions of the axiom of infinity are possible. The idea in constructing them is to postulate the existence of a set which, together with all of its elements, also contains a next “bigger” or “separate” (different) element. The other axioms can produce a potentially infinite number of different elements. However, they cannot secure the existence of a set which contains all elements which are constructed with the use of a specific method. For instance, one can build a potentially infinite series of different sets x, P(x), P(P(x)), P(P(P(x))), . . . , with the use of the Power set axiom, but the existence of a set containing all of them does not follow from the other axioms of set theory. Therefore, we need an axiom of the form: A.8. ∃z∃x ̸ = 0[(x ∈ z) ∧ ∀x(x ∈ z → P(x) ∈ z)].²¹
17 Cf. Bernays (1958). 18 A.1 is equivalent to A.2 in ZF because the only atom in ZF is the empty set 0. A quite different situation is in ZFA or in ZFC− + AFA. 19 Cf. Fraenkel, Bar-Hillel and Levy (1973), pp. 45–47. 20 For more details, see, Król (2006). 21 Obviously, when there is already another axiom of infinity, it is easy to prove A.8.
Basic Intuitions Concerning the Concept of Infinity in Mathematics |
95
One can use any method which can create a potentially infinite series of different sets. It is also possible to postulate simply that there is a potentially infinite number of such different sets. One can, or not, use the empty set or other atoms (in ZFA, i.e., ZF with atoms). However, not in every case can such an axiom be used for the construction of the set of natural numbers; see also below. Some examples of such axioms are given below: A.9. ∃z [(0 ∈ z) ∧ (∀x ∈ z ∃y ∈ z(y ̸ = 0 ∧ (x ∩ y) = 0 ∧ (x ∪ y) ∈ z)]; A.10. ∃z∃x[(x ∈ z ∧ x ̸ = 0) ∧ (∀c ∈ z ∃y ∈ z (y ̸ = 0 ∧ (c ∩ y) = 0 ∧ (c ∪ y) ∈ z)]; A.11. ∃z∃x[(0 ∈ z ∧ x ∈ z ∧ x ̸ = 0) ∧ (∀c ∈ z ∃y ∈ z (y ̸ = 0 ∧ c ̸ = y ∧ (c ∪ y) ∈ z)]; A.12. ∃z ̸ = 0[∀x ∈ z ∃y ∈ z (y ̸ = 0 ∧ (x ∩ y) = 0 ∧ (x ∪ y) ∈ z)].
The most simple axiom of infinity in this group is A.12 and it enables one to use not only the empty set x = 0. The following questions arise: is it true that every infinite set is well-founded or isomorphic to an ordinal number? Are all well-founded axioms of infinity equivalent? What are the logical relations between the concepts of actual infinity, equinumerosity and well-foundedness? Is there only one Cantor’s Paradise possible? Does every infinite set belong to Cantor’s Paradise? Some of the above axioms assert the existence of an infinite set only if the universe of sets is well-founded. For instance, if the set “z ∈ z” is not excluded then, for example in A.1 or A.12, set z can be finite for z = {x, z}. Nevertheless, the usual axioms of infinity in ZFC, such as A.2–A.3 work well also in non-wellfounded set theories; cf. for instance the (most) systems of non-well-founded set theories described in the book of P. Aczel, for instance in ZFC− +AFA.²² From the above facts one can infer that not every axiom A.1–A.12 is equivalent to the other axioms in every formal environment. Are there infinite sets which are non-well-founded? We are now going to consider some examples of infinite sets according to the above-mentioned second strategy of the construction of infinite sets. We can present some examples of such axioms. The first axiom concerns the case when the set “x ∈ x” as an element of z is not excluded: A.13. ∃z∃x(x ∈ x ∧ x ∈ z) ∧ [∀x ∈ z ∃y ∈ z (y ̸= 0 ∧ (x ∩ y) = 0 ∧ (x ∪ y) ∈ z)].
(From A.13 follows that x ̸ = z, y ̸ = x, y ̸ = z.) Obviously, there are other possible variants of this axiom depending on if the null set belongs to z or x.
22 Cf. Aczel (1988).
96 | Zbigniew Król When “x ∈ x” is excluded from z, we have, for instance (a “Mirimanoff set”): A.14. ∃z ̸ = 0 ∀x.{[x ∈ z → ¬(x ∈ x)] ∧ ∃x∀y (x ∈ y ∧ y ∈ z → x ∈ z) ∧ ∀x ∈ z ∃y (y ∈ x ∧ y ̸ = 0)}. A.14.1. ∃z ̸ = 0 {∀x, m ∈ z.¬(x ∈ x) ∧ (x ∈ m ∧ m ∈ z → x ∈ z) ∧ [∀x ∈ z ∃y (y ∈ x ∧ y ̸ = 0)]}.²³
The last two sets have no ∈-minimal elements and all the finite, as well as infinite, sequences of the form “x ∈ m ∈ u . . . ∈ x” are excluded. In these last cases, one can obtain a “two-sided” infinite set, i.e. a set which is infinite “up and down”: A.15. ∃z ̸ = 0 ∀x ∈ z.¬(x ∈ x) ∧ ∃y ∈ z (x ∈ y ∧ y ̸ = z) ∧ ∀x ∈ z ∃y ∈ z. (y ∈ x ∧ y ̸ = 0).
The last formula indicates other possible axioms in the case of well-founded sets: A.16. ∃z ̸ = 0 ∀x ∈ z ∃y ∈ z (x ∈ y ∧ y ∈ z); A.17. ∃z.0 ∈ z ∧ ∀x ∈ z ∃y ∈ z (x ∈ y ∧ y ∈ z).
The next group of non-well-founded axioms of infinity contains axioms which provides a set with an infinite circular sequence of the forms “z ∈ m ∈ u ∈ n . . . ∈ z” or “z ∈ m ∈ u ∈ n . . . ∈ u ∈ z”, etc. However, such sets are not useful for a definition of natural numbers. Such axioms as A.16 and A.17 are not necessarily transitive. On the other hand, A.14 (or A14.1) are transitive. However, they can contain many finite, as well as infinite, branches. Therefore, it is difficult to control their cardinality and to construct from them a set of natural numbers. One can easily reformulate such axioms and impose the condition of transitivity or a condition eliminating undesired branches; cf. below. The next problem is how one can enable the construction of ordinal numbers and N in a non-well-founded, as well as well-founded, case. At first, it is necessary to discern a unique specific element in such a set. We usually postulate that the null-set 0 (or an atom; cf. A.1) is an element of an infinite set. The presence of such a discerned element makes it possible to find a minimal set, i.e. a set which is contained in every infinite set.²⁴ For example, using A.2 or A.3, one cannot be sure if there are no other elements a, like 0, from which it is possible to start one more
23 Obviously, the last two axioms are equivalent. 24 One can prove, assuming the axiom schema of replacement, that the union of every family of sets is also a set.
Basic Intuitions Concerning the Concept of Infinity in Mathematics |
97
and different infinite sequence of sets, e.g. a, {a}, {{a}}, . . .. The construction of a minimal set excludes such an unwanted series. Let us consider, for instance, how one can construct a set of natural numbers with the use of A.14.1. First, it is necessary to eliminate other branches from the set z, for instance: TR. ∀x ∈ z ∀y ∈ z.x ⊂ y ∨ y ⊂ x; (∀x ∈ z ∀y ∈ z.x ∈ y ∨ y ∈ x).
TR excludes other “branches” in the set z from A.14.1 and we obtain a singular infinite “thread” of the subsets of z. We also need a condition which asserts that between any two elements x, y of z, there is only a well-founded sequence of sets m, n, k, . . . , such that x ∈ m ∈ n ∈ k ∈ . . . ∈ y. For example: Suc. ∃w ∈ z.¬(∃c ∈ z.w ∈ c) ∧ ∀x ∈ z ∃s ∈ z.(x ∈ s ∧ ¬∃m ∈ z.x ∈ m ∧ m ∈ s).
It is easy to demonstrate that such w is only one in z. Let w = Suc(z) and Suc(x ∈ w) = s ∈ z that ¬∃m ∈ z.x ∈ m ∧ m ∈ s: Suc(x)z = s iff s ∈ z ∧ x ∈ z ∧ ∀m ∈ z.¬(x ∈ m ∧ m ∈ s).
Our axiom of infinity is: A.14.1 ∧ Tr ∧ Suc (with obvious bracketing). Next, it is easy to define: 0 = z, S(w) = 1 = S(Suc(z)), n + 1 = S(S(n)) . . ., etc. One more possibility is to define an initial segment of x in z from A.14.1 ∧ Tr ∧ Suc: Ext(x)z = {s ∈ z : x ∈ s}.
For every x in z, Ext(x)z is well-founded and Ext(x)z = Ext(y)z iff x = y. Therefore, there is a sequence of ordinal numbers in Cantor’s Paradise which is isomorphic to the sequence of the well-founded sets Ext(x)z , x ∈ z, ordered by the usual relation